text-classification bool 2 classes | text stringlengths 0 664k |
|---|---|
true | # Dataset Card for kinnews_kirnews
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [More Information Needed]
- **Repository:** https://github.com/Andrews2017/KINNEWS-and-KIRNEWS-Corpus
- **Paper:** [KINNEWS and KIRNEWS: Benchmarking Cross-Lingual Text Classification for Kinyarwanda and Kirundi](https://arxiv.org/abs/2010.12174)
- **Leaderboard:** NA
- **Point of Contact:** [Rubungo Andre Niyongabo1](mailto:niyongabor.andre@std.uestc.edu.cn)
### Dataset Summary
Kinyarwanda and Kirundi news classification datasets (KINNEWS and KIRNEWS,respectively), which were both collected from Rwanda and Burundi news websites and newspapers, for low-resource monolingual and cross-lingual multiclass classification tasks.
### Supported Tasks and Leaderboards
This dataset can be used for text classification of news articles in Kinyarwadi and Kirundi languages. Each news article can be classified into one of the 14 possible classes. The classes are:
- politics
- sport
- economy
- health
- entertainment
- history
- technology
- culture
- religion
- environment
- education
- relationship
### Languages
Kinyarwanda and Kirundi
## Dataset Structure
### Data Instances
Here is an example from the dataset:
| Field | Value |
| ----- | ----------- |
| label | 1 |
| kin_label/kir_label | 'inkino' |
| url | 'https://nawe.bi/Primus-Ligue-Imirwi-igiye-guhura-gute-ku-ndwi-ya-6-y-ihiganwa.html' |
| title | 'Primus Ligue\xa0: Imirwi igiye guhura gute ku ndwi ya 6 y’ihiganwa\xa0?'|
| content | ' Inkino zitegekanijwe kuruno wa gatandatu igenekerezo rya 14 Nyakanga umwaka wa 2019...'|
| en_label| 'sport'|
### Data Fields
The raw version of the data for Kinyarwanda language consists of these fields
- label: The category of the news article
- kin_label/kir_label: The associated label in Kinyarwanda/Kirundi language
- en_label: The associated label in English
- url: The URL of the news article
- title: The title of the news article
- content: The content of the news article
The cleaned version contains only the `label`, `title` and the `content` fields
### Data Splits
Lang| Train | Test |
|---| ----- | ---- |
|Kinyarwandai Raw|17014|4254|
|Kinyarwandai Clean|17014|4254|
|Kirundi Raw|3689|923|
|Kirundi Clean|3689|923|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@article{niyongabo2020kinnews,
title={KINNEWS and KIRNEWS: Benchmarking Cross-Lingual Text Classification for Kinyarwanda and Kirundi},
author={Niyongabo, Rubungo Andre and Qu, Hong and Kreutzer, Julia and Huang, Li},
journal={arXiv preprint arXiv:2010.12174},
year={2020}
}
```
### Contributions
Thanks to [@saradhix](https://github.com/saradhix) for adding this dataset. |
false |
# Dataset Card for ChrEn
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** [Github repository for ChrEn](https://github.com/ZhangShiyue/ChrEn)
- **Paper:** [ChrEn: Cherokee-English Machine Translation for Endangered Language Revitalization](https://arxiv.org/abs/2010.04791)
- **Point of Contact:** [benfrey@email.unc.edu](benfrey@email.unc.edu)
### Dataset Summary
ChrEn is a Cherokee-English parallel dataset to facilitate machine translation research between Cherokee and English.
ChrEn is extremely low-resource contains 14k sentence pairs in total, split in ways that facilitate both in-domain and out-of-domain evaluation.
ChrEn also contains 5k Cherokee monolingual data to enable semi-supervised learning.
### Supported Tasks and Leaderboards
The dataset is intended to use for `machine-translation` between Enlish (`en`) and Cherokee (`chr`).
### Languages
The dataset contains Enlish (`en`) and Cherokee (`chr`) text. The data encompasses both existing dialects of Cherokee: the Overhill dialect, mostly spoken in Oklahoma (OK), and the Middle dialect, mostly used in North Carolina (NC).
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
Many of the source texts were translations of English materials, which means that the Cherokee structures may not be 100% natural in terms of what a speaker might spontaneously produce. Each text was translated by people who speak Cherokee as the first language, which means there is a high probability of grammaticality. These data were originally available in PDF version. We apply the Optical Character Recognition (OCR) via Tesseract OCR engine to extract the Cherokee and English text.
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
The sentences were manually aligned by Dr. Benjamin Frey a proficient second-language speaker of Cherokee, who also fixed the errors introduced by OCR. This process is time-consuming and took several months.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
The dataset was gathered and annotated by Shiyue Zhang, Benjamin Frey, and Mohit Bansal at UNC Chapel Hill.
### Licensing Information
The copyright of the data belongs to original book/article authors or translators (hence, used for research purpose; and please contact Dr. Benjamin Frey for other copyright questions).
### Citation Information
```
@inproceedings{zhang2020chren,
title={ChrEn: Cherokee-English Machine Translation for Endangered Language Revitalization},
author={Zhang, Shiyue and Frey, Benjamin and Bansal, Mohit},
booktitle={EMNLP2020},
year={2020}
}
```
### Contributions
Thanks to [@yjernite](https://github.com/yjernite), [@lhoestq](https://github.com/lhoestq) for adding this dataset. |
false |
# Dataset Card for One Billion Word Language Model Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [statmt](http://www.statmt.org/lm-benchmark/)
- **Repository:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Paper:** [arxiv](https://arxiv.org/pdf/1312.3005v3.pdf)
- **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Size of downloaded dataset files:** 1.79 GB
- **Size of the generated dataset:** 4.28 GB
- **Total amount of disk used:** 6.07 GB
### Dataset Summary
A benchmark corpus to be used for measuring progress in statistical language modeling. This has almost one billion words in the training data.
### Supported Tasks and Leaderboards
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Languages
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Dataset Structure
### Data Instances
#### plain_text
- **Size of downloaded dataset files:** 1.79 GB
- **Size of the generated dataset:** 4.28 GB
- **Total amount of disk used:** 6.07 GB
An example of 'train' looks as follows.
```
This example was too long and was cropped:
{
"text": "While athletes in different professions dealt with doping scandals and other controversies , Woods continued to do what he did best : dominate the field of professional golf and rake in endorsements ."
}
```
### Data Fields
The data fields are the same among all splits.
#### plain_text
- `text`: a `string` feature.
### Data Splits
| name | train | test |
|------------|----------|--------|
| plain_text | 30301028 | 306688 |
## Dataset Creation
### Curation Rationale
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
#### Who are the source language producers?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Annotations
The dataset doesn't contain annotations.
### Personal and Sensitive Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Discussion of Biases
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Other Known Limitations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Additional Information
### Dataset Curators
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Licensing Information
[More Information Needeate this repository accordingly.
### Citation Information
```bibtex
@misc{chelba2014billion,
title={One Billion Word Benchmark for Measuring Progress in Statistical Language Modeling},
author={Ciprian Chelba and Tomas Mikolov and Mike Schuster and Qi Ge and Thorsten Brants and Phillipp Koehn and Tony Robinson},
year={2014},
eprint={1312.3005},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
### Contributions
Thanks to [@patrickvonplaten](https://github.com/patrickvonplaten), [@lewtun](https://github.com/lewtun), [@jplu](https://github.com/jplu), [@thomwolf](https://github.com/thomwolf) for adding this dataset. |
false |
# Dataset Card for "tner/fin"
## Dataset Description
- **Repository:** [T-NER](https://github.com/asahi417/tner)
- **Paper:** [https://aclanthology.org/U15-1010.pdf](https://aclanthology.org/U15-1010.pdf)
- **Dataset:** FIN
- **Domain:** Financial News
- **Number of Entity:** 4
### Dataset Summary
FIN NER dataset formatted in a part of [TNER](https://github.com/asahi417/tner) project.
FIN dataset contains training (FIN5) and test (FIN3) only, so we randomly sample a half size of test instances from the training set to create validation set.
- Entity Types: `ORG`, `LOC`, `PER`, `MISC`
## Dataset Structure
### Data Instances
An example of `train` looks as follows.
```
{
"tags": [0, 0, 0, 0, 0, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
"tokens": ["1", ".", "1", ".", "4", "Borrower", "engages", "in", "criminal", "conduct", "or", "is", "involved", "in", "criminal", "activities", ";"]
}
```
### Label ID
The label2id dictionary can be found at [here](https://huggingface.co/datasets/tner/fin/raw/main/dataset/label.json).
```python
{
"O": 0,
"B-PER": 1,
"B-LOC": 2,
"B-ORG": 3,
"B-MISC": 4,
"I-PER": 5,
"I-LOC": 6,
"I-ORG": 7,
"I-MISC": 8
}
```
### Data Splits
| name |train|validation|test|
|---------|----:|---------:|---:|
|fin |1014 | 303| 150|
### Citation Information
```
@inproceedings{salinas-alvarado-etal-2015-domain,
title = "Domain Adaption of Named Entity Recognition to Support Credit Risk Assessment",
author = "Salinas Alvarado, Julio Cesar and
Verspoor, Karin and
Baldwin, Timothy",
booktitle = "Proceedings of the Australasian Language Technology Association Workshop 2015",
month = dec,
year = "2015",
address = "Parramatta, Australia",
url = "https://aclanthology.org/U15-1010",
pages = "84--90",
}
``` |
false |
# Dataset Card for poleval2019_mt
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** PolEval-2019 competition. http://2019.poleval.pl/
- **Repository:** Links available [in this page](http://2019.poleval.pl/index.php/tasks/task4)
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
PolEval is a SemEval-inspired evaluation campaign for natural language processing tools for Polish.
Submitted solutions compete against one another within certain tasks selected by organizers, using available data and are evaluated according to
pre-established procedures. One of the tasks in PolEval-2019 was Machine Translation (Task-4).
The task is to train as good as possible machine translation system, using any technology,with limited textual resources.
The competition will be done for 2 language pairs, more popular English-Polish (into Polish direction) and pair that can be called low resourced
Russian-Polish (in both directions).
Here, Polish-English is also made available to allow for training in both directions. However, the test data is ONLY available for English-Polish
### Supported Tasks and Leaderboards
Supports Machine Translation between Russian to Polish and English to Polish (and vice versa).
### Languages
- Polish (pl)
- Russian (ru)
- English (en)
## Dataset Structure
### Data Instances
As the training data set, a set of bi-lingual corpora aligned at the sentence level has been prepared. The corpora are saved in UTF-8 encoding as plain text, one language per file.
### Data Fields
One example of the translation is as below:
```
{
'translation': {'ru': 'не содержала в себе моделей. Модели это сравнительно новое явление. ',
'pl': 'nie miała w sobie modeli. Modele to względnie nowa dziedzina. Tak więc, jeśli '}
}
```
### Data Splits
The dataset is divided into two splits. All the headlines are scraped from news websites on the internet.
| | train | validation | test |
|-------|-------:|-----------:|-----:|
| ru-pl | 20001 | 3001 | 2969 |
| pl-ru | 20001 | 3001 | 2969 |
| en-pl | 129255 | 1000 | 9845 |
## Dataset Creation
### Curation Rationale
This data was curated as a task for the PolEval-2019. The task is to train as good as possible machine translation system, using any technology, with limited textual resources. The competition will be done for 2 language pairs, more popular English-Polish (into Polish direction) and pair that can be called low resourced Russian-Polish (in both directions).
PolEval is a SemEval-inspired evaluation campaign for natural language processing tools for Polish. Submitted tools compete against one another within certain tasks selected by organizers, using available data and are evaluated according to pre-established procedures.
PolEval 2019-related papers were presented at AI & NLP Workshop Day (Warsaw, May 31, 2019).
The links for the top performing models on various tasks (including the Task-4: Machine Translation) is present in [this](http://2019.poleval.pl/index.php/publication) link
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
The organization details of PolEval is present in this [link](http://2019.poleval.pl/index.php/organizers)
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@proceedings{ogr:kob:19:poleval,
editor = {Maciej Ogrodniczuk and Łukasz Kobyliński},
title = {{Proceedings of the PolEval 2019 Workshop}},
year = {2019},
address = {Warsaw, Poland},
publisher = {Institute of Computer Science, Polish Academy of Sciences},
url = {http://2019.poleval.pl/files/poleval2019.pdf},
isbn = "978-83-63159-28-3"}
}
```
### Contributions
Thanks to [@vrindaprabhu](https://github.com/vrindaprabhu) for adding this dataset. |
true |
# Dataset Card for Language Identification dataset
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**
- **Repository:**
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
The Language Identification dataset is a collection of 90k samples consisting of text passages and corresponding language label.
This dataset was created by collecting data from 3 sources: [Multilingual Amazon Reviews Corpus](https://huggingface.co/datasets/amazon_reviews_multi), [XNLI](https://huggingface.co/datasets/xnli), and [STSb Multi MT](https://huggingface.co/datasets/stsb_multi_mt).
### Supported Tasks and Leaderboards
The dataset can be used to train a model for language identification, which is a **multi-class text classification** task.
The model [papluca/xlm-roberta-base-language-detection](https://huggingface.co/papluca/xlm-roberta-base-language-detection), which is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base), was trained on this dataset and currently achieves 99.6% accuracy on the test set.
### Languages
The Language Identification dataset contains text in 20 languages, which are:
`arabic (ar), bulgarian (bg), german (de), modern greek (el), english (en), spanish (es), french (fr), hindi (hi), italian (it), japanese (ja), dutch (nl), polish (pl), portuguese (pt), russian (ru), swahili (sw), thai (th), turkish (tr), urdu (ur), vietnamese (vi), and chinese (zh)`
## Dataset Structure
### Data Instances
For each instance, there is a string for the text and a string for the label (the language tag). Here is an example:
`{'labels': 'fr', 'text': 'Conforme à la description, produit pratique.'}`
### Data Fields
- **labels:** a string indicating the language label.
- **text:** a string consisting of one or more sentences in one of the 20 languages listed above.
### Data Splits
The Language Identification dataset has 3 splits: *train*, *valid*, and *test*.
The train set contains 70k samples, while the validation and test sets 10k each.
All splits are perfectly balanced: the train set contains 3500 samples per language, while the validation and test sets 500.
## Dataset Creation
### Curation Rationale
This dataset was built during *The Hugging Face Course Community Event*, which took place in November 2021, with the goal of collecting a dataset with enough samples for each language to train a robust language detection model.
### Source Data
The Language Identification dataset was created by collecting data from 3 sources: [Multilingual Amazon Reviews Corpus](https://huggingface.co/datasets/amazon_reviews_multi), [XNLI](https://huggingface.co/datasets/xnli), and [STSb Multi MT](https://huggingface.co/datasets/stsb_multi_mt).
### Personal and Sensitive Information
The dataset does not contain any personal information about the authors or the crowdworkers.
## Considerations for Using the Data
### Social Impact of Dataset
This dataset was developed as a benchmark for evaluating (balanced) multi-class text classification models.
### Discussion of Biases
The possible biases correspond to those of the 3 datasets on which this dataset is based.
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@LucaPapariello](https://github.com/LucaPapariello) for adding this dataset.
|
true |
**QR-AN Dataset: a classification and generation dataset of french Parliament questions-answers.**
This is a dataset for theme/topic classification, made of questions and answers from https://www2.assemblee-nationale.fr/recherche/resultats_questions . \
It contains 188 unbalanced classes, 80k questions-answers divided into 3 splits: train (60k), val (10k) and test (10k). \
Can be used for generation with 'qran_generation'
This dataset is compatible with the [`run_summarization.py`](https://github.com/huggingface/transformers/tree/master/examples/pytorch/summarization) script from Transformers if you add this line to the `summarization_name_mapping` variable:
```python
"ccdv/cass-summarization": ("question", "answer")
```
Compatible with [run_glue.py](https://github.com/huggingface/transformers/tree/master/examples/pytorch/text-classification) script:
```
export MODEL_NAME=camembert-base
export MAX_SEQ_LENGTH=512
python run_glue.py \
--model_name_or_path $MODEL_NAME \
--dataset_name cassandra-themis/QR-AN \
--do_train \
--do_eval \
--max_seq_length $MAX_SEQ_LENGTH \
--per_device_train_batch_size 8 \
--gradient_accumulation_steps 4 \
--learning_rate 2e-5 \
--num_train_epochs 1 \
--max_eval_samples 500 \
--output_dir tmp/QR-AN
``` |
false |
# Dataset Card for "German LER"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://github.com/elenanereiss/Legal-Entity-Recognition](https://github.com/elenanereiss/Legal-Entity-Recognition)
- **Paper:** [https://arxiv.org/pdf/2003.13016v1.pdf](https://arxiv.org/pdf/2003.13016v1.pdf)
- **Point of Contact:** [elena.leitner@dfki.de](elena.leitner@dfki.de)
### Dataset Summary
A dataset of Legal Documents from German federal court decisions for Named Entity Recognition. The dataset is human-annotated with 19 fine-grained entity classes. The dataset consists of approx. 67,000 sentences and contains 54,000 annotated entities. NER tags use the `BIO` tagging scheme.
The dataset includes two different versions of annotations, one with a set of 19 fine-grained semantic classes (`ner_tags`) and another one with a set of 7 coarse-grained classes (`ner_coarse_tags`). There are 53,632 annotated entities in total, the majority of which (74.34 %) are legal entities, the others are person, location and organization (25.66 %).
For more details see [https://arxiv.org/pdf/2003.13016v1.pdf](https://arxiv.org/pdf/2003.13016v1.pdf).
### Supported Tasks and Leaderboards
- **Tasks:** Named Entity Recognition
- **Leaderboards:**
### Languages
German
## Dataset Structure
### Data Instances
```python
{
'id': '1',
'tokens': ['Eine', 'solchermaßen', 'verzögerte', 'oder', 'bewusst', 'eingesetzte', 'Verkettung', 'sachgrundloser', 'Befristungen', 'schließt', '§', '14', 'Abs.', '2', 'Satz', '2', 'TzBfG', 'aus', '.'],
'ner_tags': [38, 38, 38, 38, 38, 38, 38, 38, 38, 38, 3, 22, 22, 22, 22, 22, 22, 38, 38],
'ner_coarse_tags': [14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 2, 9, 9, 9, 9, 9, 9, 14, 14]
}
```
### Data Fields
```python
{
'id': Value(dtype='string', id=None),
'tokens': Sequence(feature=Value(dtype='string', id=None),
length=-1, id=None),
'ner_tags': Sequence(feature=ClassLabel(num_classes=39,
names=['B-AN',
'B-EUN',
'B-GRT',
'B-GS',
'B-INN',
'B-LD',
'B-LDS',
'B-LIT',
'B-MRK',
'B-ORG',
'B-PER',
'B-RR',
'B-RS',
'B-ST',
'B-STR',
'B-UN',
'B-VO',
'B-VS',
'B-VT',
'I-AN',
'I-EUN',
'I-GRT',
'I-GS',
'I-INN',
'I-LD',
'I-LDS',
'I-LIT',
'I-MRK',
'I-ORG',
'I-PER',
'I-RR',
'I-RS',
'I-ST',
'I-STR',
'I-UN',
'I-VO',
'I-VS',
'I-VT',
'O'],
id=None),
length=-1,
id=None),
'ner_coarse_tags': Sequence(feature=ClassLabel(num_classes=15,
names=['B-LIT',
'B-LOC',
'B-NRM',
'B-ORG',
'B-PER',
'B-REG',
'B-RS',
'I-LIT',
'I-LOC',
'I-NRM',
'I-ORG',
'I-PER',
'I-REG',
'I-RS',
'O'],
id=None),
length=-1,
id=None)
}
```
### Data Splits
| | train | validation | test |
|-------------------------|------:|-----------:|-----:|
| Input Sentences | 53384 | 6666 | 6673 |
## Dataset Creation
### Curation Rationale
Documents in the legal domain contain multiple references to named entities, especially domain-specific named entities, i. e., jurisdictions, legal institutions, etc. Legal documents are unique and differ greatly from newspaper texts. On the one hand, the occurrence of general-domain named entities is relatively rare. On the other hand, in concrete applications, crucial domain-specific entities need to be identified in a reliable way, such as designations of legal norms and references to other legal documents (laws, ordinances, regulations, decisions, etc.). Most NER solutions operate in the general or news domain, which makes them inapplicable to the analysis of legal documents. Accordingly, there is a great need for an NER-annotated dataset consisting of legal documents, including the corresponding development of a typology of semantic concepts and uniform annotation guidelines.
### Source Data
Court decisions from 2017 and 2018 were selected for the dataset, published online by the [Federal Ministry of Justice and Consumer Protection](http://www.rechtsprechung-im-internet.de). The documents originate from seven federal courts: Federal Labour Court (BAG), Federal Fiscal Court (BFH), Federal Court of Justice (BGH), Federal Patent Court (BPatG), Federal Social Court (BSG), Federal Constitutional Court (BVerfG) and Federal Administrative Court (BVerwG).
#### Initial Data Collection and Normalization
From the table of [contents](http://www.rechtsprechung-im-internet.de/rii-toc.xml), 107 documents from each court were selected (see Table 1). The data was collected from the XML documents, i. e., it was extracted from the XML elements `Mitwirkung, Titelzeile, Leitsatz, Tenor, Tatbestand, Entscheidungsgründe, Gründen, abweichende Meinung, and sonstiger Titel`. The metadata at the beginning of the documents (name of court, date of decision, file number, European Case Law Identifier, document type, laws) and those that belonged to previous legal proceedings was deleted. Paragraph numbers were removed.
The extracted data was split into sentences, tokenised using [SoMaJo](https://github.com/tsproisl/SoMaJo) and manually annotated in [WebAnno](https://webanno.github.io/webanno/).
#### Who are the source language producers?
The Federal Ministry of Justice and the Federal Office of Justice provide selected decisions. Court decisions were produced by humans.
### Annotations
#### Annotation process
For more details see [annotation guidelines](https://github.com/elenanereiss/Legal-Entity-Recognition/blob/master/docs/Annotationsrichtlinien.pdf) (in German).
<!-- #### Who are the annotators?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)-->
### Personal and Sensitive Information
A fundamental characteristic of the published decisions is that all personal information have been anonymised for privacy reasons. This affects the classes person, location and organization.
<!-- ## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Discussion of Biases
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Other Known Limitations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Additional Information
### Dataset Curators
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)-->
### Licensing Information
[CC BY-SA 4.0 license](https://creativecommons.org/licenses/by-sa/4.0/)
### Citation Information
```
@misc{https://doi.org/10.48550/arxiv.2003.13016,
doi = {10.48550/ARXIV.2003.13016},
url = {https://arxiv.org/abs/2003.13016},
author = {Leitner, Elena and Rehm, Georg and Moreno-Schneider, Julián},
keywords = {Computation and Language (cs.CL), Information Retrieval (cs.IR), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {A Dataset of German Legal Documents for Named Entity Recognition},
publisher = {arXiv},
year = {2020},
copyright = {arXiv.org perpetual, non-exclusive license}
}
```
### Contributions
|
false |
# Dataset Card for Never Ending Language Learning (NELL)
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**
http://rtw.ml.cmu.edu/rtw/
- **Repository:**
http://rtw.ml.cmu.edu/rtw/
- **Paper:**
Never-Ending Learning.
T. Mitchell, W. Cohen, E. Hruschka, P. Talukdar, J. Betteridge, A. Carlson, B. Dalvi, M. Gardner, B. Kisiel, J. Krishnamurthy, N. Lao, K. Mazaitis, T. Mohamed, N. Nakashole, E. Platanios, A. Ritter, M. Samadi, B. Settles, R. Wang, D. Wijaya, A. Gupta, X. Chen, A. Saparov, M. Greaves, J. Welling. In Proceedings of the Conference on Artificial Intelligence (AAAI), 2015
### Dataset Summary
This dataset provides version 1115 of the belief
extracted by CMU's Never Ending Language Learner (NELL) and version
1110 of the candidate belief extracted by NELL. See
http://rtw.ml.cmu.edu/rtw/overview. NELL is an open information
extraction system that attempts to read the Clueweb09 of 500 million
web pages (http://boston.lti.cs.cmu.edu/Data/clueweb09/) and general
web searches.
The dataset has 4 configurations: nell_belief, nell_candidate,
nell_belief_sentences, and nell_candidate_sentences. nell_belief is
certainties of belief are lower. The two sentences config extracts the
CPL sentence patterns filled with the applicable 'best' literal string
for the entities filled into the sentence patterns. And also provides
sentences found using web searches containing the entities and
relationships.
There are roughly 21M entries for nell_belief_sentences, and 100M
sentences for nell_candidate_sentences.
From the NELL website:
- **Research Goal**
To build a never-ending machine learning system that acquires the ability to extract structured information from unstructured web pages. If successful, this will result in a knowledge base (i.e., a relational database) of structured information that mirrors the content of the Web. We call this system NELL (Never-Ending Language Learner).
- **Approach**
The inputs to NELL include (1) an initial ontology defining hundreds of categories (e.g., person, sportsTeam, fruit, emotion) and relations (e.g., playsOnTeam(athlete,sportsTeam), playsInstrument(musician,instrument)) that NELL is expected to read about, and (2) 10 to 15 seed examples of each category and relation.
Given these inputs, plus a collection of 500 million web pages and access to the remainder of the web through search engine APIs, NELL runs 24 hours per day, continuously, to perform two ongoing tasks:
Extract new instances of categories and relations. In other words, find noun phrases that represent new examples of the input categories (e.g., "Barack Obama" is a person and politician), and find pairs of noun phrases that correspond to instances of the input relations (e.g., the pair "Jason Giambi" and "Yankees" is an instance of the playsOnTeam relation). These new instances are added to the growing knowledge base of structured beliefs.
Learn to read better than yesterday. NELL uses a variety of methods to extract beliefs from the web. These are retrained, using the growing knowledge base as a self-supervised collection of training examples. The result is a semi-supervised learning method that couples the training of hundreds of different extraction methods for a wide range of categories and relations. Much of NELL’s current success is due to its algorithm for coupling the simultaneous training of many extraction methods.
For more information, see: http://rtw.ml.cmu.edu/rtw/resources
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
en, and perhaps some others
## Dataset Structure
### Data Instances
There are four configurations for the dataset: nell_belief, nell_candidate, nell_belief_sentences, nell_candidate_sentences.
nell_belief and nell_candidate defines:
``
{'best_entity_literal_string': 'Aspect Medical Systems',
'best_value_literal_string': '',
'candidate_source': '%5BSEAL-Iter%3A215-2011%2F02%2F26-04%3A27%3A09-%3Ctoken%3Daspect_medical_systems%2Cbiotechcompany%3E-From%3ACategory%3Abiotechcompany-using-KB+http%3A%2F%2Fwww.unionegroup.com%2Fhealthcare%2Fmfg_info.htm+http%3A%2F%2Fwww.conventionspc.com%2Fcompanies.html%2C+CPL-Iter%3A1103-2018%2F03%2F08-15%3A32%3A34-%3Ctoken%3Daspect_medical_systems%2Cbiotechcompany%3E-grant+support+from+_%092%09research+support+from+_%094%09unrestricted+educational+grant+from+_%092%09educational+grant+from+_%092%09research+grant+support+from+_%091%09various+financial+management+positions+at+_%091%5D',
'categories_for_entity': 'concept:biotechcompany',
'categories_for_value': 'concept:company',
'entity': 'concept:biotechcompany:aspect_medical_systems',
'entity_literal_strings': '"Aspect Medical Systems" "aspect medical systems"',
'iteration_of_promotion': '1103',
'relation': 'generalizations',
'score': '0.9244426550775064',
'source': 'MBL-Iter%3A1103-2018%2F03%2F18-01%3A35%3A42-From+ErrorBasedIntegrator+%28SEAL%28aspect_medical_systems%2Cbiotechcompany%29%2C+CPL%28aspect_medical_systems%2Cbiotechcompany%29%29',
'value': 'concept:biotechcompany',
'value_literal_strings': ''}
``
nell_belief_sentences, nell_candidate_sentences defines:
``
{'count': 4,
'entity': 'biotechcompany:aspect_medical_systems',
'relation': 'generalizations',
'score': '0.9244426550775064',
'sentence': 'research support from [[ Aspect Medical Systems ]]',
'sentence_type': 'CPL',
'url': '',
'value': 'biotechcompany'}
``
### Data Fields
For nell_belief and nell_canddiate configurations. From http://rtw.ml.cmu.edu/rtw/faq:
* entity: The Entity part of the (Entity, Relation, Value) tripple. Note that this will be the name of a concept and is not the literal string of characters seen by NELL from some text source, nor does it indicate the category membership of that concept
* relation: The Relation part of the (Entity, Relation, Value) tripple. In the case of a category instance, this will be "generalizations". In the case of a relation instance, this will be the name of the relation.
* value: The Value part of the (Entity, Relation, Value) tripple. In the case of a category instance, this will be the name of the category. In the case of a relation instance, this will be another concept (like Entity).
* iteration_of_promotion: The point in NELL's life at which this category or relation instance was promoted to one that NELL beleives to be true. This is a non-negative integer indicating the number of iterations of bootstrapping NELL had gone through.
* score: A confidence score for the belief. Note that NELL's scores are not actually probabilistic at this time.
* source: A summary of the provenance for the belief indicating the set of learning subcomponents (CPL, SEAL, etc.) that had submitted this belief as being potentially true.
* entity_literal_strings: The set of actual textual strings that NELL has read that it believes can refer to the concept indicated in the Entity column.
* value_literal_strings: For relations, the set of actual textual strings that NELL has read that it believes can refer to the concept indicated in the Value column. For categories, this should be empty but may contain something spurious.
* best_entity_literal_string: Of the set of strings in the Entity literalStrings, column, which one string can best be used to describe the concept.
* best_value_literal_string: Same thing, but for Value literalStrings.
* categories_for_entity: The full set of categories (which may be empty) to which NELL belives the concept indicated in the Entity column to belong.
* categories_for_value: For relations, the full set of categories (which may be empty) to which NELL believes the concept indicated in the Value column to belong. For categories, this should be empty but may contain something spurious.
* candidate_source: A free-form amalgamation of more specific provenance information describing the justification(s) NELL has for possibly believing this category or relation instance.
For the nell_belief_sentences and nell_candidate_sentences, we have extracted the underlying sentences, sentence count and URLs and provided a shortened version of the entity, relation and value field by removing the string "concept:" and "candidate:". There are two types of sentences, 'CPL' and 'OE', which are generated by two of the modules of NELL, pattern matching and open web searching, respectively. There may be duplicates. The configuration is as follows:
* entity: The Entity part of the (Entity, Relation, Value) tripple. Note that this will be the name of a concept and is not the literal string of characters seen by NELL from some text source, nor does it indicate the category membership of that concept
* relation: The Relation part of the (Entity, Relation, Value) tripple. In the case of a category instance, this will be "generalizations". In the case of a relation instance, this will be the name of the relation.
* value: The Value part of the (Entity, Relation, Value) tripple. In the case of a category instance, this will be the name of the category. In the case of a relation instance, this will be another concept (like Entity).
* score: A confidence score for the belief. Note that NELL's scores are not actually probabilistic at this time.
* sentence: the raw sentence. For 'CPL' type sentences, there are "[[" "]]" arounds the entity and value. For 'OE' type sentences, there are no "[[" and "]]".
* url: the url if there is one from which this sentence was extracted
* count: the count for this sentence
* sentence_type: either 'CPL' or 'OE'
### Data Splits
There are no splits.
## Dataset Creation
### Curation Rationale
This dataset was gathered and created over many years of running the NELL system on web data.
### Source Data
#### Initial Data Collection and Normalization
See the research paper on NELL. NELL searches a subset of the web
(Clueweb09) and the open web using various open information extraction
algorithms, including pattern matching.
#### Who are the source language producers?
The NELL authors at Carnegie Mellon Univiersty and data from Cluebweb09 and the open web.
### Annotations
#### Annotation process
The various open information extraction modules of NELL.
#### Who are the annotators?
Machine annotated.
### Personal and Sensitive Information
Unkown, but likely there are names of famous individuals.
## Considerations for Using the Data
### Social Impact of Dataset
The goal for the work is to help machines learn to read and understand the web.
### Discussion of Biases
Since the data is gathered from the web, there is likely to be biased text and relationships.
[More Information Needed]
### Other Known Limitations
The relationships and concepts gathered from NELL are not 100% accurate, and there could be errors (maybe as high as 30% error).
See https://en.wikipedia.org/wiki/Never-Ending_Language_Learning
We did not 'tag' the entity and value in the 'OE' sentences, and this might be an extension in the future.
## Additional Information
### Dataset Curators
The authors of NELL at Carnegie Mellon Univeristy
### Licensing Information
There does not appear to be a license on http://rtw.ml.cmu.edu/rtw/resources. The data is made available by CMU on the web.
### Citation Information
@inproceedings{mitchell2015,
added-at = {2015-01-27T15:35:24.000+0100},
author = {Mitchell, T. and Cohen, W. and Hruscha, E. and Talukdar, P. and Betteridge, J. and Carlson, A. and Dalvi, B. and Gardner, M. and Kisiel, B. and Krishnamurthy, J. and Lao, N. and Mazaitis, K. and Mohammad, T. and Nakashole, N. and Platanios, E. and Ritter, A. and Samadi, M. and Settles, B. and Wang, R. and Wijaya, D. and Gupta, A. and Chen, X. and Saparov, A. and Greaves, M. and Welling, J.},
biburl = {https://www.bibsonomy.org/bibtex/263070703e6bb812852cca56574aed093/hotho},
booktitle = {AAAI},
description = {Papers by William W. Cohen},
interhash = {52d0d71f6f5b332dabc1412f18e3a93d},
intrahash = {63070703e6bb812852cca56574aed093},
keywords = {learning nell ontology semantic toread},
note = {: Never-Ending Learning in AAAI-2015},
timestamp = {2015-01-27T15:35:24.000+0100},
title = {Never-Ending Learning},
url = {http://www.cs.cmu.edu/~wcohen/pubs.html},
year = 2015
}
### Contributions
Thanks to [@ontocord](https://github.com/ontocord) for adding this dataset. |
true | ```
@article{han2022folio,
title={FOLIO: Natural Language Reasoning with First-Order Logic},
author = {Han, Simeng and Schoelkopf, Hailey and Zhao, Yilun and Qi, Zhenting and Riddell, Martin and Benson, Luke and Sun, Lucy and Zubova, Ekaterina and Qiao, Yujie and Burtell, Matthew and Peng, David and Fan, Jonathan and Liu, Yixin and Wong, Brian and Sailor, Malcolm and Ni, Ansong and Nan, Linyong and Kasai, Jungo and Yu, Tao and Zhang, Rui and Joty, Shafiq and Fabbri, Alexander R. and Kryscinski, Wojciech and Lin, Xi Victoria and Xiong, Caiming and Radev, Dragomir},
journal={arXiv preprint arXiv:2209.00840},
url = {https://arxiv.org/abs/2209.00840},
year={2022}
``` |
true |
# Dataset Card for Greek Legal Code
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** https://github.com/christospi/glc-nllp-21
- **Paper:** https://arxiv.org/abs/2109.15298
- **Data:** https://doi.org/10.5281/zenodo.5528002
- **Leaderboard:** N/A
- **Point of Contact:** [Christos Papaloukas](mailto:christospap@di.uoa.gr)
### Dataset Summary
Greek_Legal_Code (GLC) is a dataset consisting of approx. 47k legal resources from Greek legislation. The origin of GLC is “Permanent Greek Legislation Code - Raptarchis”, a collection of Greek legislative documents classified into multi-level (from broader to more specialized) categories.
**Topics**
GLC consists of 47 legislative volumes and each volume corresponds to a main thematic topic. Each volume is divided into thematic sub categories which are called chapters and subsequently, each chapter breaks down to subjects which contain the legal resources. The total number of chapters is 389 while the total number of subjects is 2285, creating an interlinked thematic hierarchy. So, for the upper thematic level (volume) GLC has 47 classes. For the next thematic level (chapter) GLC offers 389 classes and for the inner and last thematic level (subject), GLC has 2285 classes.
GLC classes are divided into three categories for each thematic level: frequent classes, which occur in more than 10 training documents and can be found in all three subsets (training, development and test); few-shot classes which appear in 1 to 10 training documents and also appear in the documents of the development and test sets, and zero-shot classes which appear in the development and/or test, but not in the training documents.
### Supported Tasks and Leaderboards
The dataset supports:
**Multi-class Text Classification:** Given the text of a document, a model predicts the corresponding class.
**Few-shot and Zero-shot learning:** As already noted, the classes can be divided into three groups: frequent, few-shot, and zero- shot, depending on whether they were assigned to more than 10, fewer than 10 but at least one, or no training documents, respectively.
| Level | Total | Frequent | Few-Shot (<10) | Zero-Shot |
|---|---|---|---|---|
|Volume|47|47|0|0|
|Chapter|389|333|53|3|
|Subject|2285|712|1431|142|
### Languages
All documents are written in Greek.
## Dataset Structure
### Data Instances
```json
{
"text": "179. ΑΠΟΦΑΣΗ ΥΠΟΥΡΓΟΥ ΜΕΤΑΦΟΡΩΝ ΚΑΙ ΕΠΙΚΟΙΝΩΝΙΩΝ Αριθ. Β-οικ. 68425/4765 της 2/17 Νοεμ. 2000 (ΦΕΚ Β΄ 1404) Τροποποίηση της 42000/2030/81 κοιν. απόφασης του Υπουργού Συγκοινωνιών «Κωδικοποίηση και συμπλήρωση καν. Αποφάσεων» που εκδόθηκαν κατ’ εξουσιοδότηση του Ν.Δ. 102/73 «περί οργανώσεως των δια λεωφορείων αυτοκινήτων εκτελουμένων επιβατικών συγκοινωνιών». ",
"volume": 24, # "ΣΥΓΚΟΙΝΩΝΙΕΣ"
}
```
### Data Fields
The following data fields are provided for documents (`train`, `dev`, `test`):
`text`: (**str**) The full content of each document, which is represented by its `header` and `articles` (i.e., the `main_body`).\
`label`: (**class label**): Depending on the configurarion, the volume/chapter/subject of the document. For volume-level class it belongs to specifically: ["ΚΟΙΝΩΝΙΚΗ ΠΡΟΝΟΙΑ",
"ΓΕΩΡΓΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΡΑΔΙΟΦΩΝΙΑ ΚΑΙ ΤΥΠΟΣ",
"ΒΙΟΜΗΧΑΝΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΥΓΕΙΟΝΟΜΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΠΟΛΕΜΙΚΟ ΝΑΥΤΙΚΟ",
"ΤΑΧΥΔΡΟΜΕΙΑ - ΤΗΛΕΠΙΚΟΙΝΩΝΙΕΣ",
"ΔΑΣΗ ΚΑΙ ΚΤΗΝΟΤΡΟΦΙΑ",
"ΕΛΕΓΚΤΙΚΟ ΣΥΝΕΔΡΙΟ ΚΑΙ ΣΥΝΤΑΞΕΙΣ",
"ΠΟΛΕΜΙΚΗ ΑΕΡΟΠΟΡΙΑ",
"ΝΟΜΙΚΑ ΠΡΟΣΩΠΑ ΔΗΜΟΣΙΟΥ ΔΙΚΑΙΟΥ",
"ΝΟΜΟΘΕΣΙΑ ΑΝΩΝΥΜΩΝ ΕΤΑΙΡΕΙΩΝ ΤΡΑΠΕΖΩΝ ΚΑΙ ΧΡΗΜΑΤΙΣΤΗΡΙΩΝ",
"ΠΟΛΙΤΙΚΗ ΑΕΡΟΠΟΡΙΑ",
"ΕΜΜΕΣΗ ΦΟΡΟΛΟΓΙΑ",
"ΚΟΙΝΩΝΙΚΕΣ ΑΣΦΑΛΙΣΕΙΣ",
"ΝΟΜΟΘΕΣΙΑ ΔΗΜΩΝ ΚΑΙ ΚΟΙΝΟΤΗΤΩΝ",
"ΝΟΜΟΘΕΣΙΑ ΕΠΙΜΕΛΗΤΗΡΙΩΝ ΣΥΝΕΤΑΙΡΙΣΜΩΝ ΚΑΙ ΣΩΜΑΤΕΙΩΝ",
"ΔΗΜΟΣΙΑ ΕΡΓΑ",
"ΔΙΟΙΚΗΣΗ ΔΙΚΑΙΟΣΥΝΗΣ",
"ΑΣΦΑΛΙΣΤΙΚΑ ΤΑΜΕΙΑ",
"ΕΚΚΛΗΣΙΑΣΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΕΚΠΑΙΔΕΥΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΔΗΜΟΣΙΟ ΛΟΓΙΣΤΙΚΟ",
"ΤΕΛΩΝΕΙΑΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΣΥΓΚΟΙΝΩΝΙΕΣ",
"ΕΘΝΙΚΗ ΑΜΥΝΑ",
"ΣΤΡΑΤΟΣ ΞΗΡΑΣ",
"ΑΓΟΡΑΝΟΜΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΔΗΜΟΣΙΟΙ ΥΠΑΛΛΗΛΟΙ",
"ΠΕΡΙΟΥΣΙΑ ΔΗΜΟΣΙΟΥ ΚΑΙ ΝΟΜΙΣΜΑ",
"ΟΙΚΟΝΟΜΙΚΗ ΔΙΟΙΚΗΣΗ",
"ΛΙΜΕΝΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΑΣΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΠΟΛΙΤΙΚΗ ΔΙΚΟΝΟΜΙΑ",
"ΔΙΠΛΩΜΑΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΔΙΟΙΚΗΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΑΜΕΣΗ ΦΟΡΟΛΟΓΙΑ",
"ΤΥΠΟΣ ΚΑΙ ΤΟΥΡΙΣΜΟΣ",
"ΕΘΝΙΚΗ ΟΙΚΟΝΟΜΙΑ",
"ΑΣΤΥΝΟΜΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΑΓΡΟΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΕΡΓΑΤΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΠΟΙΝΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΕΜΠΟΡΙΚΗ ΝΟΜΟΘΕΣΙΑ",
"ΕΠΙΣΤΗΜΕΣ ΚΑΙ ΤΕΧΝΕΣ",
"ΕΜΠΟΡΙΚΗ ΝΑΥΤΙΛΙΑ",
"ΣΥΝΤΑΓΜΑΤΙΚΗ ΝΟΜΟΘΕΣΙΑ"
] \
The labels can also be a the chapter-level or subject-level class it belongs to. Some chapter labels are omitted due to size (389 classes). Some subject labels are also omitted due to size (2285 classes).
### Data Splits
| Split | No of Documents | Avg. words |
| ------------------- | ------------------------------------ | --- |
| Train | 28,536 | 600 |
|Development | 9,511 | 574 |
|Test | 9,516 | 595 |
## Dataset Creation
### Curation Rationale
The dataset was curated by Papaloukas et al. (2021) with the hope to support and encourage further research in NLP for the Greek language.
### Source Data
#### Initial Data Collection and Normalization
The ``Permanent Greek Legislation Code - Raptarchis`` is a thorough catalogue of Greek legislation since the creation of the Greek state in 1834 until 2015. It includes Laws, Royal and Presidential Decrees, Regulations and Decisions, retrieved from the Official Government Gazette, where Greek legislation is published. This collection is one of the official, publicly available sources of classified Greek legislation suitable for classification tasks.
Currently, the original catalogue is publicly offered in MS Word (.doc) format through the portal e-Themis, the legal database and management service of it, under the administration of the Ministry of the Interior (Affairs). E-Themis is primarily focused on providing legislation on a multitude of predefined thematic categories, as described in the catalogue. The main goal is to help users find legislation of interest using the thematic index.
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
The dataset does not include personal or sensitive information.
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Papaloukas et al. (2021)
### Licensing Information
[More Information Needed]
### Citation Information
*Christos Papaloukas, Ilias Chalkidis, Konstantinos Athinaios, Despina-Athanasia Pantazi and Manolis Koubarakis.*
*Multi-granular Legal Topic Classification on Greek Legislation.*
*Proceedings of the 3rd Natural Legal Language Processing (NLLP) Workshop, Punta Cana, Dominican Republic, 2021*
```
@inproceedings{papaloukas-etal-2021-glc,
title = "Multi-granular Legal Topic Classification on Greek Legislation",
author = "Papaloukas, Christos and Chalkidis, Ilias and Athinaios, Konstantinos and Pantazi, Despina-Athanasia and Koubarakis, Manolis",
booktitle = "Proceedings of the Natural Legal Language Processing Workshop 2021",
year = "2021",
address = "Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/2109.15298",
doi = "10.48550/arXiv.2109.15298",
pages = "63--75"
}
```
### Contributions
Thanks to [@christospi](https://github.com/christospi) for adding this dataset. |
false |
# Dataset Card for "qanta"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [http://www.qanta.org/](http://www.qanta.org/)
- **Repository:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Paper:** [Quizbowl: The Case for Incremental Question Answering](https://arxiv.org/abs/1904.04792)
- **Point of Contact:** [Jordan Boyd-Graber](mailto:jbg@umiacs.umd.edu)
- **Size of downloaded dataset files:** 170.75 MB
- **Size of the generated dataset:** 147.18 MB
- **Total amount of disk used:** 317.93 MB
### Dataset Summary
The Qanta dataset is a question answering dataset based on the academic trivia game Quizbowl.
### Supported Tasks and Leaderboards
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Languages
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Dataset Structure
### Data Instances
#### mode=first,char_skip=25
- **Size of downloaded dataset files:** 170.75 MB
- **Size of the generated dataset:** 147.18 MB
- **Total amount of disk used:** 317.93 MB
An example of 'guessdev' looks as follows.
```
This example was too long and was cropped:
{
"answer": "Apollo_program",
"category": "History",
"char_idx": -1,
"dataset": "quizdb.org",
"difficulty": "easy_college",
"first_sentence": "As part of this program, William Anders took a photo that Galen Rowell called \"the most influential environmental photograph ever taken.\"",
"fold": "guessdev",
"full_question": "\"As part of this program, William Anders took a photo that Galen Rowell called \\\"the most influential environmental photograph e...",
"gameplay": false,
"id": "127028-first",
"page": "Apollo_program",
"proto_id": "",
"qanta_id": 127028,
"qdb_id": 126689,
"raw_answer": "Apollo program [or Project Apollo; accept Apollo 8; accept Apollo 1; accept Apollo 11; prompt on landing on the moon]",
"sentence_idx": -1,
"subcategory": "American",
"text": "As part of this program, William Anders took a photo that Galen Rowell called \"the most influential environmental photograph ever taken.\"",
"tokenizations": [[0, 137], [138, 281], [282, 412], [413, 592], [593, 675]],
"tournament": "ACF Fall",
"year": 2016
}
```
### Data Fields
The data fields are the same among all splits.
#### mode=first,char_skip=25
- `id`: a `string` feature.
- `qanta_id`: a `int32` feature.
- `proto_id`: a `string` feature.
- `qdb_id`: a `int32` feature.
- `dataset`: a `string` feature.
- `text`: a `string` feature.
- `full_question`: a `string` feature.
- `first_sentence`: a `string` feature.
- `char_idx`: a `int32` feature.
- `sentence_idx`: a `int32` feature.
- `tokenizations`: a dictionary feature containing:
- `feature`: a `int32` feature.
- `answer`: a `string` feature.
- `page`: a `string` feature.
- `raw_answer`: a `string` feature.
- `fold`: a `string` feature.
- `gameplay`: a `bool` feature.
- `category`: a `string` feature.
- `subcategory`: a `string` feature.
- `tournament`: a `string` feature.
- `difficulty`: a `string` feature.
- `year`: a `int32` feature.
### Data Splits
| name |adversarial|buzzdev|buzztrain|guessdev|guesstrain|buzztest|guesstest|
|-----------------------|----------:|------:|--------:|-------:|---------:|-------:|--------:|
|mode=first,char_skip=25| 1145| 1161| 16706| 1055| 96221| 1953| 2151|
## Dataset Creation
### Curation Rationale
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
#### Who are the source language producers?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Annotations
#### Annotation process
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
#### Who are the annotators?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Personal and Sensitive Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Discussion of Biases
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Other Known Limitations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Additional Information
### Dataset Curators
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Licensing Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Citation Information
```
@article{Rodriguez2019QuizbowlTC,
title={Quizbowl: The Case for Incremental Question Answering},
author={Pedro Rodriguez and Shi Feng and Mohit Iyyer and He He and Jordan L. Boyd-Graber},
journal={ArXiv},
year={2019},
volume={abs/1904.04792}
}
```
### Contributions
Thanks to [@thomwolf](https://github.com/thomwolf), [@patrickvonplaten](https://github.com/patrickvonplaten), [@lewtun](https://github.com/lewtun) for adding this dataset. |
false |
# Dataset Card for Dataset Name
## Dataset Description
- **Homepage:**
- **Repository:**
https://github.com/uci-soe/FairytaleQAData
https://github.com/WorkInTheDark/FairytaleQA_Dataset
- **Paper:**
https://aclanthology.org/2022.acl-long.34/
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
This is the repository for the FairytaleQA dataset, an open-source dataset focusing on comprehension of narratives, targeting students from kindergarten to eighth grade. The FairytaleQA dataset is annotated by education experts based on an evidence-based theoretical framework. It consists of 10,580 explicit and implicit questions derived from 278 children-friendly stories, covering seven types of narrative elements or relations.
### Supported Tasks and Leaderboards
Question-Answering, Question-Generation, Question-Answer Pair Generation
### Languages
English
## Dataset Structure
### Data Instances
An example of "train" looks as follows:
```
{
'story_name': 'three-dogs',
'story_section': 'once upon a time there was a king who went forth into the world and
... ...
guards to watch over the little princess so that she would not get out under the open sky .',
'question': 'why was there great rejoicing in the city and throughout the country ?',
'answer1': 'the people wished their king all that was good .',
'answer2': '',
'local-or-sum': 'local',
'attribute': 'causal relationship',
'ex-or-im': 'explicit',
'ex-or-im2': '',
}
```
### Data Fields
- **'story_name'**: story name
- **'story_section'**: story section related to the QA-pair
- **'question'**: the question content
- **'answer1'**: the 1st answer (available in all splits)
- **'answer2'**: the 2nd answer by another annotator (only available in test / val splits)
- **'local-or-sum'**: 'local' denotes the question is related to only one story section, while 'summary' denotes the question is related to multiple story sections
- **'attribute'**: categorized by education experts into seven narrative elements: character / setting / action / feeling / causal relationship / outcome resolution, detailed definition is described in the paper
- **'ex-or-im'**: 'explicit' denotes the answer can be found in the story content, while 'implicit' denotes the answer require high-level summarization
- **'ex-or-im2'**: similar to 'ex-or-im', but annotated by another annotator (only available in storys in test / val splits)
### Data Splits
- train split: 232 books with 8548 QA-pairs
- val split: 23 books with 1025 QA-pairs
- test split: 23 books with 1007 QA-pairs
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
Our Dataset Paper is accepted to ACL 2022, you may cite:
```
@inproceedings{xu2022fairytaleqa,
author={Xu, Ying and Wang, Dakuo and Yu, Mo and Ritchie, Daniel and Yao, Bingsheng and Wu, Tongshuang and Zhang, Zheng and Li, Toby Jia-Jun and Bradford, Nora and Sun, Branda and Hoang, Tran Bao and Sang, Yisi and Hou, Yufang and Ma, Xiaojuan and Yang, Diyi and Peng, Nanyun and Yu, Zhou and Warschauer, Mark},
title = {Fantastic Questions and Where to Find Them: Fairytale{QA} -- An Authentic Dataset for Narrative Comprehension},
publisher = {Association for Computational Linguistics},
year = {2022}
}
```
### Contributions
[More Information Needed] |
true |
# Dataset Card for HateCheck
## Dataset Description
HateCheck is a suite of functional test for hate speech detection models.
The dataset contains 3,728 validated test cases in 29 functional tests.
19 functional tests correspond to distinct types of hate. The other 11 functional tests cover challenging types of non-hate.
This allows for targeted diagnostic insights into model performance.
In our ACL paper, we found critical weaknesses in all commercial and academic hate speech detection model that we tested with HateCheck.
Please refer to the paper (linked below) for results and further discussion, as well as further information on the dataset and a full data statement.
- **Paper:** Röttger et al. (2021) - HateCheck: Functional Tests for Hate Speech Detection Model. https://aclanthology.org/2021.acl-long.4/ or https://arxiv.org/abs/2012.15606
- **Repository:** https://github.com/paul-rottger/hatecheck-data
- **Point of Contact:** paul.rottger@oii.ox.ac.uk
## Dataset Structure
"test.csv" contains all 3,728 validated test cases. Each test case (row) has the following attributes:
**functionality**
The shorthand for the functionality tested by the test case.
**case_id**
The unique ID of the test case (assigned to each of the 3,901 cases we initially generated)
**test_case**
The text of the test case.
**label_gold**
The gold standard label (hateful/non-hateful) of the test case. All test cases within a given functionality have the same gold standard label.
**target_ident**
Where applicable, the protected group targeted or referenced by the test case. We cover seven protected groups in the test suite: women, trans people, gay people, black people, disabled people, Muslims and immigrants.
**direction**
For hateful cases, the binary secondary label indicating whether they are *directed* at an individual as part of a protected group or aimed at the group in *general*.
**focus_words**
Where applicable, the key word or phrase in a given test case (e.g. "cut their throats").
**focus_lemma**
Where applicable, the corresponding lemma (e.g. "cut sb. throat").
**ref_case_id**
For hateful cases, where applicable, the ID of the simpler hateful case which was perturbed to generate them.
For non-hateful cases, where applicable, the ID of the hateful case which is contrasted.
**ref_templ_id**
The equivalent, but for template IDs.
**templ_id**
The unique ID of the template from which the test case was generated (assigned to each of the 866 cases and templates from which we generated the 3,901 initial cases).
## Citation Information
When using HateCheck, please cite our ACL paper:
@inproceedings{rottger-etal-2021-hatecheck,
title = "{H}ate{C}heck: Functional Tests for Hate Speech Detection Models",
author = {R{\"o}ttger, Paul and
Vidgen, Bertie and
Nguyen, Dong and
Waseem, Zeerak and
Margetts, Helen and
Pierrehumbert, Janet},
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.4",
doi = "10.18653/v1/2021.acl-long.4",
pages = "41--58",
abstract = "Detecting online hate is a difficult task that even state-of-the-art models struggle with. Typically, hate speech detection models are evaluated by measuring their performance on held-out test data using metrics such as accuracy and F1 score. However, this approach makes it difficult to identify specific model weak points. It also risks overestimating generalisable model performance due to increasingly well-evidenced systematic gaps and biases in hate speech datasets. To enable more targeted diagnostic insights, we introduce HateCheck, a suite of functional tests for hate speech detection models. We specify 29 model functionalities motivated by a review of previous research and a series of interviews with civil society stakeholders. We craft test cases for each functionality and validate their quality through a structured annotation process. To illustrate HateCheck{'}s utility, we test near-state-of-the-art transformer models as well as two popular commercial models, revealing critical model weaknesses.",
}
|
false |
# Dataset Card for GEM/mlsum
## Dataset Description
- **Homepage:** N/A
- **Repository:** https://gitlab.lip6.fr/scialom/mlsum_data/-/tree/master/MLSUM
- **Paper:** https://aclanthology.org/2020.emnlp-main.647/
- **Leaderboard:** N/A
- **Point of Contact:** Thomas Scialom
### Link to Main Data Card
You can find the main data card on the [GEM Website](https://gem-benchmark.com/data_cards/mlsum).
### Dataset Summary
MLSum is a multilingual summarization dataset crawled from different news websites. The GEM version supports the German and Spanish subset alongside specifically collected challenge sets for COVID-related articles to test out-of-domain generalization.
You can load the dataset via:
```
import datasets
data = datasets.load_dataset('GEM/mlsum')
```
The data loader can be found [here](https://huggingface.co/datasets/GEM/mlsum).
#### website
N/A
#### paper
[ACL Anthology](https://aclanthology.org/2020.emnlp-main.647/)
#### authors
Thomas Scialom, Paul-Alexis Dray, Sylvain Lamprier, Benjamin Piwowarski, Jacopo Staiano
## Dataset Overview
### Where to find the Data and its Documentation
#### Download
<!-- info: What is the link to where the original dataset is hosted? -->
<!-- scope: telescope -->
[Gitlab](https://gitlab.lip6.fr/scialom/mlsum_data/-/tree/master/MLSUM)
#### Paper
<!-- info: What is the link to the paper describing the dataset (open access preferred)? -->
<!-- scope: telescope -->
[ACL Anthology](https://aclanthology.org/2020.emnlp-main.647/)
#### BibTex
<!-- info: Provide the BibTex-formatted reference for the dataset. Please use the correct published version (ACL anthology, etc.) instead of google scholar created Bibtex. -->
<!-- scope: microscope -->
```
@inproceedings{scialom-etal-2020-mlsum,
title = "{MLSUM}: The Multilingual Summarization Corpus",
author = "Scialom, Thomas and
Dray, Paul-Alexis and
Lamprier, Sylvain and
Piwowarski, Benjamin and
Staiano, Jacopo",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.emnlp-main.647",
doi = "10.18653/v1/2020.emnlp-main.647",
pages = "8051--8067",
abstract = "We present MLSUM, the first large-scale MultiLingual SUMmarization dataset. Obtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages {--} namely, French, German, Spanish, Russian, Turkish. Together with English news articles from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community. We report cross-lingual comparative analyses based on state-of-the-art systems. These highlight existing biases which motivate the use of a multi-lingual dataset.",
}
```
#### Contact Name
<!-- quick -->
<!-- info: If known, provide the name of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
Thomas Scialom
#### Contact Email
<!-- info: If known, provide the email of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
{thomas,paul-alexis,jacopo}@recital.ai, {sylvain.lamprier,benjamin.piwowarski}@lip6.fr
#### Has a Leaderboard?
<!-- info: Does the dataset have an active leaderboard? -->
<!-- scope: telescope -->
no
### Languages and Intended Use
#### Multilingual?
<!-- quick -->
<!-- info: Is the dataset multilingual? -->
<!-- scope: telescope -->
yes
#### Covered Dialects
<!-- info: What dialects are covered? Are there multiple dialects per language? -->
<!-- scope: periscope -->
There is only one dialect per language, Hochdeutsch for German and Castilian Spanish for Spanish.
#### Covered Languages
<!-- quick -->
<!-- info: What languages/dialects are covered in the dataset? -->
<!-- scope: telescope -->
`German`, `Spanish, Castilian`
#### Whose Language?
<!-- info: Whose language is in the dataset? -->
<!-- scope: periscope -->
The German articles are crawled from Süddeutsche Zeitung and the Spanish ones from El Pais.
#### License
<!-- quick -->
<!-- info: What is the license of the dataset? -->
<!-- scope: telescope -->
other: Other license
#### Intended Use
<!-- info: What is the intended use of the dataset? -->
<!-- scope: microscope -->
The intended use of this dataset is to augment existing datasets for English news summarization with additional languages.
#### Add. License Info
<!-- info: What is the 'other' license of the dataset? -->
<!-- scope: periscope -->
Restricted to non-commercial research purposes.
#### Primary Task
<!-- info: What primary task does the dataset support? -->
<!-- scope: telescope -->
Summarization
#### Communicative Goal
<!-- quick -->
<!-- info: Provide a short description of the communicative goal of a model trained for this task on this dataset. -->
<!-- scope: periscope -->
The speaker is required to produce a high quality summary of news articles in the same language as the input article.
### Credit
#### Curation Organization Type(s)
<!-- info: In what kind of organization did the dataset curation happen? -->
<!-- scope: telescope -->
`other`
#### Curation Organization(s)
<!-- info: Name the organization(s). -->
<!-- scope: periscope -->
CNRS, Sorbonne Université, reciTAL
#### Dataset Creators
<!-- info: Who created the original dataset? List the people involved in collecting the dataset and their affiliation(s). -->
<!-- scope: microscope -->
Thomas Scialom, Paul-Alexis Dray, Sylvain Lamprier, Benjamin Piwowarski, Jacopo Staiano
#### Funding
<!-- info: Who funded the data creation? -->
<!-- scope: microscope -->
Funding information is not specified.
#### Who added the Dataset to GEM?
<!-- info: Who contributed to the data card and adding the dataset to GEM? List the people+affiliations involved in creating this data card and who helped integrate this dataset into GEM. -->
<!-- scope: microscope -->
The original data card was written by Pedro Henrique Martins (Instituto de Telecomunicações) and Sebastian Gehrmann (Google Research) extended and updated it to the v2 format. The COVID challenge set was created by Laura Perez-Beltrachini (University of Edinburgh). Data cleaning was done by Juan Diego Rodriguez (UT Austin).
### Dataset Structure
#### Data Fields
<!-- info: List and describe the fields present in the dataset. -->
<!-- scope: telescope -->
The data fields are:
- `text`: the source article (`string`).
- `summary`: the output summary (`string`).
- `topic`: the topic of the article (`string`).
- `url`: the article's url (`string`).
- `title`: the article's title (`string`).
- `date`: the article's date (`string`).
#### Reason for Structure
<!-- info: How was the dataset structure determined? -->
<!-- scope: microscope -->
The structure follows previously released datasets. The `topic` and `title` fields were added to enable additional tasks like title generation and topic detection.
#### How were labels chosen?
<!-- info: How were the labels chosen? -->
<!-- scope: microscope -->
They are human written highlights or summaries scraped from the same website.
#### Example Instance
<!-- info: Provide a JSON formatted example of a typical instance in the dataset. -->
<!-- scope: periscope -->
```
{
'date': '00/01/2010',
'gem_id': 'mlsum_de-train-2',
'gem_parent_id': 'mlsum_de-train-2',
'references': [],
'target': 'Oskar Lafontaine gibt den Parteivorsitz der Linken ab - und seine Kollegen streiten, wer ihn beerben soll. sueddeutsche.de stellt die derzeit aussichtsreichsten Anwärter für Führungsaufgaben vor. Mit Vote.',
'text': 'Wenn an diesem Montag die Landesvorsitzenden der Linken über die Nachfolger der derzeitigen Chefs Lothar Bisky und Oskar Lafontaine sowie des Bundesgeschäftsführers Dietmar Bartsch beraten, geht es nicht nur darum, wer die Partei führen soll. Es geht auch um die künftige Ausrichtung und Stärke einer Partei, die vor allem von Lafontaine zusammengehalten worden war. Ihm war es schließlich vor fünf Jahren gelungen, aus der ostdeutschen PDS und der westedeutschen WASG eine Partei zu formen. Eine Partei allerdings, die zerrissen ist in Ost und West, in Regierungswillige und ewige Oppositionelle, in Realos und Ideologen, in gemäßigte und radikale Linke. Wir stellen mögliche Kandidaten vor. Stimmen Sie ab: Wen halten Sie für geeignet und wen für unfähig? Kampf um Lafontaines Erbe: Gregor Gysi Sollte überhaupt jemand die Partei alleine führen, wie es sich viele Ostdeutsche wünschen, käme dafür wohl nur der 62-jährige Gregor Gysi in Betracht. Er ist nach Lafontaine einer der bekanntesten Politiker der Linken und derzeit Fraktionsvorsitzender der Partei im Bundestag. Allerdings ist der ehemalige PDS-Vorsitzende und Rechtsanwalt nach drei Herzinfarkten gesundheitlich angeschlagen. Wahrscheinlich wäre deshalb, dass er die zerstrittene Partei nur übergangsweise führt. Doch noch ist nicht klar, ob eine Person allein die Partei führen soll oder eine Doppelspitze. Viele Linke wünschen sich ein Duo aus einem westdeutschen und einem ostdeutschen Politiker, Mann und Frau. Foto: Getty Images',
'title': 'Personaldebatte bei der Linken - Wer kommt nach Lafontaine?',
'topic': 'politik',
'url': 'https://www.sueddeutsche.de/politik/personaldebatte-bei-der-linken-wer-kommt-nach-lafontaine-1.70041'
}
```
#### Data Splits
<!-- info: Describe and name the splits in the dataset if there are more than one. -->
<!-- scope: periscope -->
The statistics of the original dataset are:
| | Dataset | Train | Validation | Test | Mean article length | Mean summary length |
| :--- | :----: | :---: | :---: | :---: | :---: | :---: |
| German | 242,982 | 220,887 |11,394 |10,701 |570.6 (words) | 30.36 (words) |
| Spanish | 290,645 | 266,367 |10,358 |13,920 |800.5 (words) |20.71 (words) |
The statistics of the cleaned version of the dataset are:
| | Dataset | Train | Validation | Test |
| :--- | :----: | :---: | :---: | :---: |
| German | 242,835 | 220,887 |11,392 |10,695 |
| Spanish | 283,228 |259,886 |9,977 |13,365 |
The COVID challenge sets have 5058 (de) and 1938 (es) examples.
#### Splitting Criteria
<!-- info: Describe any criteria for splitting the data, if used. If there are differences between the splits (e.g., if the training annotations are machine-generated and the dev and test ones are created by humans, or if different numbers of annotators contributed to each example), describe them here. -->
<!-- scope: microscope -->
The training set contains data from 2010 to 2018. Data from 2019 (~10% of the dataset) is used for validation (up to May) and testing(May-December 2019).
####
<!-- info: What does an outlier of the dataset in terms of length/perplexity/embedding look like? -->
<!-- scope: microscope -->
Some topics are less represented within the dataset (e.g., Financial news in German and Television in Spanish).
## Dataset in GEM
### Rationale for Inclusion in GEM
#### Why is the Dataset in GEM?
<!-- info: What does this dataset contribute toward better generation evaluation and why is it part of GEM? -->
<!-- scope: microscope -->
As the first large-scale multilingual summarization dataset, it enables evaluation of summarization models beyond English.
#### Similar Datasets
<!-- info: Do other datasets for the high level task exist? -->
<!-- scope: telescope -->
yes
#### Unique Language Coverage
<!-- info: Does this dataset cover other languages than other datasets for the same task? -->
<!-- scope: periscope -->
yes
#### Difference from other GEM datasets
<!-- info: What else sets this dataset apart from other similar datasets in GEM? -->
<!-- scope: microscope -->
In our configuration, the dataset is fully non-English.
#### Ability that the Dataset measures
<!-- info: What aspect of model ability can be measured with this dataset? -->
<!-- scope: periscope -->
Content Selection, Content Planning, Realization
### GEM-Specific Curation
#### Modificatied for GEM?
<!-- info: Has the GEM version of the dataset been modified in any way (data, processing, splits) from the original curated data? -->
<!-- scope: telescope -->
yes
#### GEM Modifications
<!-- info: What changes have been made to he original dataset? -->
<!-- scope: periscope -->
`data points removed`, `data points added`
#### Modification Details
<!-- info: For each of these changes, described them in more details and provided the intended purpose of the modification -->
<!-- scope: microscope -->
The modifications done to the original dataset are the following:
- Selection of 2 languages (Spanish and German) out of the dataset 5 languages due to copyright restrictions.
- Removal of duplicate articles.
- Manually removal of article-summary pairs for which the summary is not related to the article.
- Removal of article-summary pairs written in a different language (detected using the [langdetect](https://pypi.org/project/langdetect/) library).
#### Additional Splits?
<!-- info: Does GEM provide additional splits to the dataset? -->
<!-- scope: telescope -->
yes
#### Split Information
<!-- info: Describe how the new splits were created -->
<!-- scope: periscope -->
For both selected languages (German and Spanish), we compiled time-shifted test data in the form of new articles for the second semester of 2020 with Covid19-related keywords. We collected articles from the same German and Spanish outlets as the original MLSUM datasets (El Pais and Süddeutsche Zeitung). We used the scripts provided for the re-creation of the [MLSUM datasets](https://github.com/recitalAI/MLSUM). The new challenge test set for German contains 5058 instances and the Spanish one contains 1938.
We additionally sample 500 training and validation points as additional challenge sets to measure overfitting.
#### Split Motivation
<!-- info: What aspects of the model's generation capacities were the splits created to test? -->
<!-- scope: periscope -->
Generalization to unseen topics.
### Getting Started with the Task
## Previous Results
### Previous Results
#### Measured Model Abilities
<!-- info: What aspect of model ability can be measured with this dataset? -->
<!-- scope: telescope -->
Content Selection, Content Planning, Realization
#### Metrics
<!-- info: What metrics are typically used for this task? -->
<!-- scope: periscope -->
`METEOR`, `ROUGE`, `Other: Other Metrics`
#### Other Metrics
<!-- info: Definitions of other metrics -->
<!-- scope: periscope -->
Novelty: Number of generated n-grams not included in the source articles.
#### Proposed Evaluation
<!-- info: List and describe the purpose of the metrics and evaluation methodology (including human evaluation) that the dataset creators used when introducing this task. -->
<!-- scope: microscope -->
ROUGE and METEOR both measure n-gram overlap with a focus on recall and are standard summarization metrics. Novelty is often reported alongside them to characterize how much a model diverges from its inputs.
#### Previous results available?
<!-- info: Are previous results available? -->
<!-- scope: telescope -->
yes
#### Other Evaluation Approaches
<!-- info: What evaluation approaches have others used? -->
<!-- scope: periscope -->
The GEM benchmark results (https://gem-benchmark.com/results) report a wide range of metrics include lexical overlap metrics but also semantic ones like BLEURT and BERT-Score.
## Dataset Curation
### Original Curation
#### Original Curation Rationale
<!-- info: Original curation rationale -->
<!-- scope: telescope -->
The rationale was to create a multilingual news summarization dataset that mirrors the format of popular English datasets like XSum or CNN/DM.
#### Communicative Goal
<!-- info: What was the communicative goal? -->
<!-- scope: periscope -->
The speaker is required to produce a high quality summary of news articles in the same language as the input article.
#### Sourced from Different Sources
<!-- info: Is the dataset aggregated from different data sources? -->
<!-- scope: telescope -->
yes
#### Source Details
<!-- info: List the sources (one per line) -->
<!-- scope: periscope -->
www.lemonde.fr
www.sueddeutsche.de
www.elpais.com
www.mk.ru
www.internethaber.com
### Language Data
#### How was Language Data Obtained?
<!-- info: How was the language data obtained? -->
<!-- scope: telescope -->
`Found`
#### Where was it found?
<!-- info: If found, where from? -->
<!-- scope: telescope -->
`Multiple websites`
#### Language Producers
<!-- info: What further information do we have on the language producers? -->
<!-- scope: microscope -->
The language producers are professional journalists.
#### Topics Covered
<!-- info: Does the language in the dataset focus on specific topics? How would you describe them? -->
<!-- scope: periscope -->
4/5 of the original languages report their topics (except Turkish) and the distributions differ between sources. The dominant topics in German are Politik, Sport, Wirtschaft (economy). The dominant topics in Spanish are actualidad (current news) and opinion. French and Russian are different as well but we omit these languages in the GEM version.
#### Data Validation
<!-- info: Was the text validated by a different worker or a data curator? -->
<!-- scope: telescope -->
not validated
#### Was Data Filtered?
<!-- info: Were text instances selected or filtered? -->
<!-- scope: telescope -->
algorithmically
#### Filter Criteria
<!-- info: What were the selection criteria? -->
<!-- scope: microscope -->
In the original dataset, only one filter was applied: all the articles shorter than 50 words or summaries shorter than 10 words are discarded.
The GEM version additionally applies langID filter to ensure that articles are in the correct language.
### Structured Annotations
#### Additional Annotations?
<!-- quick -->
<!-- info: Does the dataset have additional annotations for each instance? -->
<!-- scope: telescope -->
none
#### Annotation Service?
<!-- info: Was an annotation service used? -->
<!-- scope: telescope -->
no
### Consent
#### Any Consent Policy?
<!-- info: Was there a consent policy involved when gathering the data? -->
<!-- scope: telescope -->
no
#### Justification for Using the Data
<!-- info: If not, what is the justification for reusing the data? -->
<!-- scope: microscope -->
The copyright remains with the original data creators and the usage permission is restricted to non-commercial uses.
### Private Identifying Information (PII)
#### Contains PII?
<!-- quick -->
<!-- info: Does the source language data likely contain Personal Identifying Information about the data creators or subjects? -->
<!-- scope: telescope -->
yes/very likely
#### Categories of PII
<!-- info: What categories of PII are present or suspected in the data? -->
<!-- scope: periscope -->
`sensitive information`, `generic PII`
#### Any PII Identification?
<!-- info: Did the curators use any automatic/manual method to identify PII in the dataset? -->
<!-- scope: periscope -->
no identification
### Maintenance
#### Any Maintenance Plan?
<!-- info: Does the original dataset have a maintenance plan? -->
<!-- scope: telescope -->
no
## Broader Social Context
### Previous Work on the Social Impact of the Dataset
#### Usage of Models based on the Data
<!-- info: Are you aware of cases where models trained on the task featured in this dataset ore related tasks have been used in automated systems? -->
<!-- scope: telescope -->
no
### Impact on Under-Served Communities
#### Addresses needs of underserved Communities?
<!-- info: Does this dataset address the needs of communities that are traditionally underserved in language technology, and particularly language generation technology? Communities may be underserved for exemple because their language, language variety, or social or geographical context is underepresented in NLP and NLG resources (datasets and models). -->
<!-- scope: telescope -->
no
### Discussion of Biases
#### Any Documented Social Biases?
<!-- info: Are there documented social biases in the dataset? Biases in this context are variations in the ways members of different social categories are represented that can have harmful downstream consequences for members of the more disadvantaged group. -->
<!-- scope: telescope -->
no
|
false |
# Dataset Card for leNER-br
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [leNER-BR homepage](https://cic.unb.br/~teodecampos/LeNER-Br/)
- **Repository:** [leNER-BR repository](https://github.com/peluz/lener-br)
- **Paper:** [leNER-BR: Long Form Question Answering](https://cic.unb.br/~teodecampos/LeNER-Br/luz_etal_propor2018.pdf)
- **Point of Contact:** [Pedro H. Luz de Araujo](mailto:pedrohluzaraujo@gmail.com)
### Dataset Summary
LeNER-Br is a Portuguese language dataset for named entity recognition
applied to legal documents. LeNER-Br consists entirely of manually annotated
legislation and legal cases texts and contains tags for persons, locations,
time entities, organizations, legislation and legal cases.
To compose the dataset, 66 legal documents from several Brazilian Courts were
collected. Courts of superior and state levels were considered, such as Supremo
Tribunal Federal, Superior Tribunal de Justiça, Tribunal de Justiça de Minas
Gerais and Tribunal de Contas da União. In addition, four legislation documents
were collected, such as "Lei Maria da Penha", giving a total of 70 documents
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
The language supported is Portuguese.
## Dataset Structure
### Data Instances
An example from the dataset looks as follows:
```
{
"id": "0",
"ner_tags": [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 0, 0],
"tokens": [
"EMENTA", ":", "APELAÇÃO", "CÍVEL", "-", "AÇÃO", "DE", "INDENIZAÇÃO", "POR", "DANOS", "MORAIS", "-", "PRELIMINAR", "-", "ARGUIDA", "PELO", "MINISTÉRIO", "PÚBLICO", "EM", "GRAU", "RECURSAL"]
}
```
### Data Fields
- `id`: id of the sample
- `tokens`: the tokens of the example text
- `ner_tags`: the NER tags of each token
The NER tags correspond to this list:
```
"O", "B-ORGANIZACAO", "I-ORGANIZACAO", "B-PESSOA", "I-PESSOA", "B-TEMPO", "I-TEMPO", "B-LOCAL", "I-LOCAL", "B-LEGISLACAO", "I-LEGISLACAO", "B-JURISPRUDENCIA", "I-JURISPRUDENCIA"
```
The NER tags have the same format as in the CoNLL shared task: a B denotes the first item of a phrase and an I any non-initial word.
### Data Splits
The data is split into train, validation and test set. The split sizes are as follow:
| Train | Val | Test |
| ------ | ----- | ---- |
| 7828 | 1177 | 1390 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{luz_etal_propor2018,
author = {Pedro H. {Luz de Araujo} and Te\'{o}filo E. {de Campos} and
Renato R. R. {de Oliveira} and Matheus Stauffer and
Samuel Couto and Paulo Bermejo},
title = {{LeNER-Br}: a Dataset for Named Entity Recognition in {Brazilian} Legal Text},
booktitle = {International Conference on the Computational Processing of Portuguese ({PROPOR})},
publisher = {Springer},
series = {Lecture Notes on Computer Science ({LNCS})},
pages = {313--323},
year = {2018},
month = {September 24-26},
address = {Canela, RS, Brazil},
doi = {10.1007/978-3-319-99722-3_32},
url = {https://cic.unb.br/~teodecampos/LeNER-Br/},
}
```
### Contributions
Thanks to [@jonatasgrosman](https://github.com/jonatasgrosman) for adding this dataset. |
false |
## 内容
包含约200万条由[BELLE](https://github.com/LianjiaTech/BELLE)项目生成的中文指令数据。
## 样例
```
{
"instruction": "将以下三个句子组合成一个有意义的段落。\n狗是人类最好的朋友。它们非常聪明，可以进行各种活动。如果你喜欢散步，狗可以成为你一起散步的伙伴。",
"input": "",
"output": "狗是人类最好的朋友，它们非常聪明，可以进行各种活动。如果你喜欢散步，狗可以成为你一起散步的伙伴。出门散步是一种良好的锻炼方式，而有狗的陪伴会让散步变得更有趣，并且有狗在身边也能给你带来安全感。所以，拥有一只狗作为你的伙伴，可以帮助你变得更加积极主动和健康。"
}
```
### 字段：
```
instruction: 指令
input: 输入（本数据集均为空）
output: 输出
```
## 使用限制
仅允许将此数据集及使用此数据集生成的衍生物用于研究目的，不得用于商业，以及其他会对社会带来危害的用途。
本数据集不代表任何一方的立场、利益或想法，无关任何团体的任何类型的主张。因使用本数据集带来的任何损害、纠纷，本项目不承担任何责任。 |
false |
# Dataset Card for JNLPBA
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** http://www.geniaproject.org/shared-tasks/bionlp-jnlpba-shared-task-2004
- **Repository:** [Needs More Information]
- **Paper:** https://www.aclweb.org/anthology/W04-1213.pdf
- **Leaderboard:** https://paperswithcode.com/sota/named-entity-recognition-ner-on-jnlpba?p=biobert-a-pre-trained-biomedical-language
- **Point of Contact:** [Needs More Information]
### Dataset Summary
The data came from the GENIA version 3.02 corpus (Kim et al., 2003). This was formed from a controlled search on MEDLINE using the MeSH terms human, blood cells and transcription factors. From this search 2,000 abstracts were selected and hand annotated according to a small taxonomy of 48 classes based on a chemical classification. Among the classes, 36 terminal classes were used to annotate the GENIA corpus.
### Supported Tasks and Leaderboards
NER
### Languages
English
## Dataset Structure
### Data Instances
{
'id': '1',
'tokens': ['IL-2', 'gene', 'expression', 'and', 'NF-kappa', 'B', 'activation', 'through', 'CD28', 'requires', 'reactive', 'oxygen', 'production', 'by', '5-lipoxygenase', '.'],
'ner_tags': [1, 2, 0, 0, 9, 10, 0, 0, 9, 0, 0, 0, 0, 0, 9, 0],
}
### Data Fields
- `id`: Sentence identifier.
- `tokens`: Array of tokens composing a sentence.
- `ner_tags`: Array of tags, where `0` indicates no bio-entity mentioned, `1` signals the first token of a bio-entity and `2` the subsequent bio-entity tokens.
### Data Splits
Train samples: 37094
Validation samples: 7714
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
@inproceedings{collier-kim-2004-introduction,
title = "Introduction to the Bio-entity Recognition Task at {JNLPBA}",
author = "Collier, Nigel and
Kim, Jin-Dong",
booktitle = "Proceedings of the International Joint Workshop on Natural Language Processing in Biomedicine and its Applications ({NLPBA}/{B}io{NLP})",
month = aug # " 28th and 29th",
year = "2004",
address = "Geneva, Switzerland",
publisher = "COLING",
url = "https://aclanthology.org/W04-1213",
pages = "73--78",
}
### Contributions
Thanks to [@edugp](https://github.com/edugp) for adding this dataset. |
false |
# Dataset Card for "code_x_glue_cc_code_refinement"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits-sample-size)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/code-refinement
### Dataset Summary
CodeXGLUE code-refinement dataset, available at https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/code-refinement
We use the dataset released by this paper(https://arxiv.org/pdf/1812.08693.pdf). The source side is a Java function with bugs and the target side is the refined one. All the function and variable names are normalized. Their dataset contains two subsets ( i.e.small and medium) based on the function length.
### Supported Tasks and Leaderboards
- `text2text-generation-other-debugging`: The dataset can be used to train a model for automatically fixing buggy code.
### Languages
- Java **programming** language
## Dataset Structure
### Data Instances
#### medium
An example of 'train' looks as follows.
```
{
"buggy": "public static TYPE_1 init ( java.lang.String name , java.util.Date date ) { TYPE_1 VAR_1 = new TYPE_1 ( ) ; VAR_1 . METHOD_1 ( name ) ; java.util.Calendar VAR_2 = java.util.Calendar.getInstance ( ) ; VAR_2 . METHOD_2 ( date ) ; VAR_1 . METHOD_3 ( VAR_2 ) ; return VAR_1 ; }\n",
"fixed": "public static TYPE_1 init ( java.lang.String name , java.util.Date date ) { TYPE_1 VAR_1 = new TYPE_1 ( ) ; VAR_1 . METHOD_1 ( name ) ; java.util.Calendar VAR_2 = null ; if ( date != null ) { VAR_2 = java.util.Calendar.getInstance ( ) ; VAR_2 . METHOD_2 ( date ) ; } VAR_1 . METHOD_3 ( VAR_2 ) ; return VAR_1 ; }\n",
"id": 0
}
```
#### small
An example of 'validation' looks as follows.
```
{
"buggy": "public java.util.List < TYPE_1 > METHOD_1 ( ) { java.util.ArrayList < TYPE_1 > VAR_1 = new java.util.ArrayList < TYPE_1 > ( ) ; for ( TYPE_2 VAR_2 : VAR_3 ) { VAR_1 . METHOD_2 ( VAR_2 . METHOD_1 ( ) ) ; } return VAR_1 ; } \n",
"fixed": "public java.util.List < TYPE_1 > METHOD_1 ( ) { return VAR_1 ; } \n",
"id": 0
}
```
### Data Fields
In the following each data field in go is explained for each config. The data fields are the same among all splits.
#### medium, small
|field name| type | description |
|----------|------|--------------------------------|
|id |int32 | Index of the sample |
|buggy |string| The buggy version of the code |
|fixed |string| The correct version of the code|
### Data Splits
| name |train|validation|test|
|------|----:|---------:|---:|
|medium|52364| 6546|6545|
|small |46680| 5835|5835|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
Downloaded from GitHub Archive every public GitHub event between March 2011 and October 2017 and used the Google BigQuery APIs.
[More Information Needed]
#### Who are the source language producers?
Software Engineering developers.
### Annotations
#### Annotation process
Automatically annotated by filtering commit messages containing the pattern: ("fix" or "solve") and ("bug" or "issue" or "problem" or "error"). A statistically significant amount of samples (95% confidence level with 5% confidence interval) were manually evaluated by two authors to check if the filtered bug/fix pairs were correct. After all disagreements were settled, authors conclude that 97.6% were true positives.
#### Who are the annotators?
Heuristics and the authors of the paper.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
https://github.com/microsoft, https://github.com/madlag
### Licensing Information
Computational Use of Data Agreement (C-UDA) License.
### Citation Information
```
@article{CodeXGLUE,
title={CodeXGLUE: A Benchmark Dataset and Open Challenge for Code Intelligence},
year={2020},}
```
### Contributions
Thanks to @madlag (and partly also @ncoop57) for adding this dataset. |
false |
# Dataset Card for GermaNER
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** https://github.com/tudarmstadt-lt/GermaNER
- **Paper:** https://pdfs.semanticscholar.org/b250/3144ed2152830f6c64a9f797ab3c5a34fee5.pdf
- **Point of Contact:** [Darina Benikova](mailto:benikova@aiphes.tu-darmstadt.de)
### Dataset Summary
[More Information Needed]
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
German
## Dataset Structure
### Data Instances
An example instance looks as follows:
```
{
'id': '3',
'ner_tags': [1, 5, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8],
'tokens': ['Bayern', 'München', 'ist', 'wieder', 'alleiniger', 'Top-', 'Favorit', 'auf', 'den', 'Gewinn', 'der', 'deutschen', 'Fußball-Meisterschaft', '.']
}
```
### Data Fields
Each instance in the dataset has:
- `id`: an id as a string
- `tokens`: sequence of tokens
- `ner_tags`: NER tags for each token (encoded as IOB)
NER tags can be: 'B-LOC' (0), 'B-ORG' (1), 'B-OTH' (2), 'B-PER' (3), 'I-LOC' (4), 'I-ORG' (5), 'I-OTH' (6), 'I-PER' (7), 'O' (8)
### Data Splits
Dataset provides only train part (26200 data instances).
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
[More Information Needed]
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
[More Information Needed]
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
License of GermaNER:
```
GermaNER is licensed under ASL 2.0 and other lenient licenses, allowing its use for academic and commercial purposes without restrictions. The licenses of its compenents are mixed licensed and are individually listed in Data/Licenses.
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction, and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all other entities that control, are controlled by, or are under common control with that entity. For the purposes of this definition, "control" means (i) the power, direct or indirect, to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications, including but not limited to software source code, documentation source, and configuration files.
"Object" form shall mean any form resulting from mechanical transformation or translation of a Source form, including but not limited to compiled object code, generated documentation, and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or Object form, made available under the License, as indicated by a copyright notice that is included in or attached to the work (an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object form, that is based on (or derived from) the Work and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. For the purposes of this License, Derivative Works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including the original version of the Work and any modifications or additions to that Work or Derivative Works thereof, that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner. For the purposes of this definition, "submitted" means any form of electronic, verbal, or written communication sent to the Licensor or its representatives, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Work, but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by combination of their Contribution(s) with the Work to which such Contribution(s) was submitted. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work or a Contribution incorporated within the Work constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the Work or Derivative Works thereof in any medium, with or without modifications, and in Source or Object form, provided that You meet the following conditions:
You must give any other recipients of the Work or Derivative Works a copy of this License; and
You must cause any modified files to carry prominent notices stating that You changed the files; and
You must retain, in the Source form of any Derivative Works that You distribute, all copyright, patent, trademark, and attribution notices from the Source form of the Work, excluding those notices that do not pertain to any part of the Derivative Works; and
If the Work includes a "NOTICE" text file as part of its distribution, then any Derivative Works that You distribute must include a readable copy of the attribution notices contained within such NOTICE file, excluding those notices that do not pertain to any part of the Derivative Works, in at least one of the following places: within a NOTICE text file distributed as part of the Derivative Works; within the Source form or documentation, if provided along with the Derivative Works; or, within a display generated by the Derivative Works, if and wherever such third-party notices normally appear. The contents of the NOTICE file are for informational purposes only and do not modify the License. You may add Your own attribution notices within Derivative Works that You distribute, alongside or as an addendum to the NOTICE text from the Work, provided that such additional attribution notices cannot be construed as modifying the License.
You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such Derivative Works as a whole, provided Your use, reproduction, and distribution of the Work otherwise complies with the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise, any Contribution intentionally submitted for inclusion in the Work by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions. Notwithstanding the above, nothing herein shall supersede or modify the terms of any separate license agreement you may have executed with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor, except as required for reasonable and customary use in describing the origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or agreed to in writing, Licensor provides the Work (and each Contributor provides its Contributions) on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied, including, without limitation, any warranties or conditions of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE. You are solely responsible for determining the appropriateness of using or redistributing the Work and assume any risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory, whether in tort (including negligence), contract, or otherwise, unless required by applicable law (such as deliberate and grossly negligent acts) or agreed to in writing, shall any Contributor be liable to You for damages, including any direct, indirect, special, incidental, or consequential damages of any character arising as a result of this License or out of the use or inability to use the Work (including but not limited to damages for loss of goodwill, work stoppage, computer failure or malfunction, or any and all other commercial damages or losses), even if such Contributor has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing the Work or Derivative Works thereof, You may choose to offer, and charge a fee for, acceptance of support, warranty, indemnity, or other liability obligations and/or rights consistent with this License. However, in accepting such obligations, You may act only on Your own behalf and on Your sole responsibility, not on behalf of any other Contributor, and only if You agree to indemnify, defend, and hold each Contributor harmless for any liability incurred by, or claims asserted against, such Contributor by reason of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
```
### Citation Information
```bibtex
@inproceedings{Benikova2015GermaNERFO,
title={GermaNER: Free Open German Named Entity Recognition Tool},
author={Darina Benikova and Seid Muhie Yimam and P. Santhanam and Chris Biemann},
booktitle={GSCL},
year={2015}
}
```
### Contributions
Thanks to [@abhishekkrthakur](https://github.com/abhishekkrthakur) for adding this dataset. |
true |
# Dataset Card for [Dataset Name]
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Hope Speech Detection for Equality, Diversity, and Inclusion-EACL 2021](https://competitions.codalab.org/competitions/27653#learn_the_details)
- **Repository:** [HopeEDI data repository](https://competitions.codalab.org/competitions/27653#participate-get_data)
- **Paper:** [HopeEDI: A Multilingual Hope Speech Detection Dataset for Equality, Diversity, and Inclusion](https://www.aclweb.org/anthology/2020.peoples-1.5/)
- **Leaderboard:** [Rank list](https://competitions.codalab.org/competitions/27653#results)
- **Point of Contact:** [Bharathi Raja Chakravarthi](mailto:bharathiraja.akr@gmail.com)
### Dataset Summary
A Hope Speech dataset for Equality, Diversity and Inclusion (HopeEDI) containing user-generated comments from the social media platform YouTube with 28,451, 20,198 and 10,705 comments in English, Tamil and Malayalam, respectively, manually labelled as containing hope speech or not. To our knowledge, this is the first research of its kind to annotate hope speech for equality, diversity and inclusion in a multilingual setting.
### Supported Tasks and Leaderboards
To identify hope speech in the comments/posts in social media.
### Languages
English, Tamil and Malayalam
## Dataset Structure
### Data Instances
An example from the English dataset looks as follows:
| text | label |
| :------ | :----- |
| all lives matter .without that we never have peace so to me forever all lives matter. | Hope_speech |
| I think it's cool that you give people a voice to speak out with here on this channel. | Hope_speech |
An example from the Tamil dataset looks as follows:
| text | label |
| :------ | :----- |
| Idha solla ivalo naala | Non_hope_speech |
| இன்று தேசிய பெண் குழந்தைகள் தினம்.. பெண் குழந்தைகளை போற்றுவோம்..அவர்களை பாதுகாப்போம்... | Hope_speech |
An example from the Malayalam dataset looks as follows:
| text | label |
| :------ | :----- |
| ഇത്രെയും കഷ്ടപ്പെട്ട് വളർത്തിയ ആ അമ്മയുടെ മുഖം കണ്ടപ്പോൾ കണ്ണ് നിറഞ്ഞു പോയി | Hope_speech |
| snehikunavar aanayalum pennayalum onnichu jeevikatte..aareyum compel cheythitallalooo..parasparamulla ishtathodeyalle...avarum jeevikatte..🥰🥰 | Hope_speech |
### Data Fields
English
- `text`: English comment.
- `label`: list of the possible values: "Hope_speech", "Non_hope_speech", "not-English"
Tamil
- `text`: Tamil-English code mixed comment.
- `label`: list of the possible values: "Hope_speech", "Non_hope_speech", "not-Tamil"
Malayalam
- `text`: Malayalam-English code mixed comment.
- `label`: list of the possible values: "Hope_speech", "Non_hope_speech", "not-malayalam"
### Data Splits
| | train | validation |
| ----- |------:|-----------:|
| English | 22762 | 2843 |
| Tamil | 16160 | 2018 |
| Malayalam | 8564 | 1070 |
## Dataset Creation
### Curation Rationale
Hope is considered significant for the well-being, recuperation and restoration of human life by health professionals.
Hate speech or offensive language detection dataset is not available for code-mixed Tamil and code-mixed Malayalam, and it does not take into account LGBTIQ, women in STEM and other minorities. Thus, we cannot use existing hate speech or offensive language detection datasets to detect hope or non-hope for EDI of minorities.
### Source Data
#### Initial Data Collection and Normalization
For English, we collected data on recent topics of EDI, including women in STEM, LGBTIQ issues, COVID-19, Black Lives Matters, United Kingdom (UK) versus China, United States of America (USA) versus China and Australia versus China from YouTube video comments. The data was collected from videos of people from English-speaking countries, such as Australia, Canada, the Republic of Ireland, United Kingdom, the United States of America and New Zealand.
For Tamil and Malayalam, we collected data from India on the recent topics regarding LGBTIQ issues, COVID-19, women in STEM, the Indo-China war and Dravidian affairs.
#### Who are the source language producers?
Youtube users
### Annotations
#### Annotation process
We created Google forms to collect annotations from annotators. Each form contained a maximum of 100 comments, and each page contained a maximum of 10 comments to maintain the quality of annotation. We collected information on the gender, educational background and the medium of schooling of the annotator to know the diversity of the annotator and avoid bias. We educated annotators by providing them with YouTube videos on EDI. A minimum of three annotators annotated each form.
#### Who are the annotators?
For English language comments, annotators were from Australia, the Republic of Ireland, the United Kingdom and the United States of America. For Tamil, we were able to get annotations from both people from the state of Tamil Nadu of India and from Sri Lanka. Most of the annotators were graduate or post-graduate students.
### Personal and Sensitive Information
Social media data is highly sensitive, and even more so when it is related to the minority population, such as the LGBTIQ community or women. We have taken full consideration to minimise the risk associated with individual identity in the data by removing personal information from dataset, such as names but not celebrity names. However, to study EDI, we needed to keep information relating to the following characteristics; racial, gender, sexual orientation, ethnic origin and philosophical beliefs. Annotators were only shown anonymised posts and agreed to make no attempts to contact the comment creator. The dataset will only be made available for research purpose to the researcher who agree to follow ethical
guidelines
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
This work is licensed under a [Creative Commons Attribution 4.0 International Licence](http://creativecommons.org/licenses/by/4.0/.)
### Citation Information
```
@inproceedings{chakravarthi-2020-hopeedi,
title = "{H}ope{EDI}: A Multilingual Hope Speech Detection Dataset for Equality, Diversity, and Inclusion",
author = "Chakravarthi, Bharathi Raja",
booktitle = "Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.peoples-1.5",
pages = "41--53",
abstract = "Over the past few years, systems have been developed to control online content and eliminate abusive, offensive or hate speech content. However, people in power sometimes misuse this form of censorship to obstruct the democratic right of freedom of speech. Therefore, it is imperative that research should take a positive reinforcement approach towards online content that is encouraging, positive and supportive contents. Until now, most studies have focused on solving this problem of negativity in the English language, though the problem is much more than just harmful content. Furthermore, it is multilingual as well. Thus, we have constructed a Hope Speech dataset for Equality, Diversity and Inclusion (HopeEDI) containing user-generated comments from the social media platform YouTube with 28,451, 20,198 and 10,705 comments in English, Tamil and Malayalam, respectively, manually labelled as containing hope speech or not. To our knowledge, this is the first research of its kind to annotate hope speech for equality, diversity and inclusion in a multilingual setting. We determined that the inter-annotator agreement of our dataset using Krippendorff{'}s alpha. Further, we created several baselines to benchmark the resulting dataset and the results have been expressed using precision, recall and F1-score. The dataset is publicly available for the research community. We hope that this resource will spur further research on encouraging inclusive and responsive speech that reinforces positiveness.",
}
```
### Contributions
Thanks to [@jamespaultg](https://github.com/jamespaultg) for adding this dataset. |
false |
# Dataset Card for `scb_mt_enth_2020`
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://airesearch.in.th/
- **Repository:** https://github.com/vistec-AI/thai2nmt
- **Paper:** https://arxiv.org/abs/2007.03541
- **Leaderboard:**
- **Point of Contact:** https://airesearch.in.th/
### Dataset Summary
scb-mt-en-th-2020: A Large English-Thai Parallel Corpus
The primary objective of our work is to build a large-scale English-Thai dataset for machine translation.
We construct an English-Thai machine translation dataset with over 1 million segment pairs, curated from various sources,
namely news, Wikipedia articles, SMS messages, task-based dialogs, web-crawled data and government documents.
Methodology for gathering data, building parallel texts and removing noisy sentence pairs are presented in a reproducible manner.
We train machine translation models based on this dataset. Our models' performance are comparable to that of
Google Translation API (as of May 2020) for Thai-English and outperform Google when the Open Parallel Corpus (OPUS) is
included in the training data for both Thai-English and English-Thai translation.
The dataset, pre-trained models, and source code to reproduce our work are available for public use.
### Supported Tasks and Leaderboards
machine translation
### Languages
English, Thai
## Dataset Structure
### Data Instances
```
{'subdataset': 'aqdf', 'translation': {'en': 'FAR LEFT: Indonesian National Police Chief Tito Karnavian, from left, Philippine National Police Chief Ronald Dela Rosa and Royal Malaysian Police Inspector General Khalid Abu Bakar link arms before the Trilateral Security Meeting in Pasay city, southeast of Manila, Philippines, in June 2017. [THE ASSOCIATED PRESS]', 'th': '(ซ้ายสุด) นายติโต คาร์นาเวียน ผู้บัญชาการตํารวจแห่งชาติอินโดนีเซีย (จากซ้าย) นายโรนัลด์ เดลา โรซา ผู้บัญชาการตํารวจแห่งชาติฟิลิปปินส์ และนายคาลิด อาบู บาการ์ ผู้บัญชาการตํารวจแห่งชาติมาเลเซีย ไขว้แขนกันก่อนเริ่มการประชุมความมั่นคงไตรภาคีในเมืองปาเซย์ ซึ่งอยู่ทางตะวันออกเฉียงใต้ของกรุงมะนิลา ประเทศฟิลิปปินส์ ในเดือนมิถุนายน พ.ศ. 2560 ดิแอสโซซิเอทเต็ด เพรส'}}
{'subdataset': 'thai_websites', 'translation': {'en': "*Applicants from certain countries may be required to pay a visa issuance fee after their application is approved. The Department of State's website has more information about visa issuance fees and can help you determine if an issuance fee applies to your nationality.", 'th': 'ประเภทวีซ่า รวมถึงค่าธรรมเนียม และข้อกําหนดในการสัมภาษณ์วีซ่า จะขึ้นอยู่กับชนิดของหนังสือเดินทาง และจุดประสงค์ในการเดินทางของท่าน โปรดดูตารางด้านล่างก่อนการสมัครวีซ่า'}}
{'subdataset': 'nus_sms', 'translation': {'en': 'Yup... Okay. Cya tmr... So long nvr write already... Dunno whether tmr can come up with 500 words', 'th': 'ใช่...ได้ แล้วเจอกันพรุ่งนี้... นานแล้วไม่เคยเขียน... ไม่รู้ว่าพรุ่งนี้จะทําได้ถึง500คําไหมเลย'}}
```
### Data Fields
- `subdataset`: subdataset from which the sentence pair comes from
- `translation`:
- `en`: English sentences (original source)
- `th`: Thai sentences (originally target for translation)
### Data Splits
```
Split ratio (train, valid, test) : (0.8, 0.1, 0.1)
Number of paris (train, valid, test): 801,402 | 100,173 | 100,177
# Train
generated_reviews_yn: 218,637 ( 27.28% )
task_master_1: 185,671 ( 23.17% )
generated_reviews_translator: 105,561 ( 13.17% )
thai_websites: 93,518 ( 11.67% )
paracrawl: 46,802 ( 5.84% )
nus_sms: 34,495 ( 4.30% )
mozilla_common_voice: 2,451 ( 4.05% )
wikipedia: 26,163 ( 3.26% cd)
generated_reviews_crowd: 19,769 ( 2.47% )
assorted_government: 19,712 ( 2.46% )
aqdf: 10,466 ( 1.31% )
msr_paraphrase: 8,157 ( 1.02% )
# Valid
generated_reviews_yn: 30,786 ( 30.73% )
task_master_1: 18,531 ( 18.50% )
generated_reviews_translator: 13,884 ( 13.86% )
thai_websites: 13,381 ( 13.36% )
paracrawl: 6,618 ( 6.61% )
nus_sms: 4,628 ( 4.62% )
wikipedia: 3,796 ( 3.79% )
assorted_government: 2,842 ( 2.83% )
generated_reviews_crowd: 2,409 ( 2.40% )
aqdf: 1,518 ( 1.52% )
msr_paraphrase: 1,107 ( 1.11% )
mozilla_common_voice: 673 ( 0.67% )
# Test
generated_reviews_yn: 30,785 ( 30.73% )
task_master_1: 18,531 ( 18.50% )
generated_reviews_translator: 13,885 ( 13.86% )
thai_websites: 13,381 ( 13.36% )
paracrawl: 6,619 ( 6.61% )
nus_sms: 4,627 ( 4.62% )
wikipedia: 3,797 ( 3.79% )
assorted_government: 2,844 ( 2.83% )
generated_reviews_crowd: 2,409 ( 2.40% )
aqdf: 1,519 ( 1.52% )
msr_paraphrase: 1,107 ( 1.11% )
mozilla_common_voice : 673 ( 0.67% )
```
## Dataset Creation
### Curation Rationale
[AIResearch](https://airesearch.in.th/), funded by [VISTEC](https://www.vistec.ac.th/) and [depa](https://www.depa.or.th/th/home), curated this dataset as part of public NLP infrastructure. The center releases the dataset and baseline models under CC-BY-SA 4.0.
### Source Data
#### Initial Data Collection and Normalization
The sentence pairs are curated from news, Wikipedia articles, SMS messages, task-based dialogs, webcrawled data and government documents. Sentence pairs are generated by:
- Professional translators
- Crowdsourced translators
- Google Translate API and human annotators (accepted or rejected)
- Sentence alignment with [multilingual universal sentence encoder](https://tfhub.dev/google/universal-sentence-encoder-multilingual/3); the author created [CRFCut](https://github.com/vistec-AI/crfcut) to segment Thai sentences to be abel to align with their English counterparts (sentence segmented by [NLTK](https://www.nltk.org/))
For detailed explanation of dataset curation, see https://arxiv.org/pdf/2007.03541.pdf
### Annotations
#### Sources and Annotation process
- generated_reviews_yn: generated by [CTRL](https://arxiv.org/abs/1909.05858), translated to Thai by Google Translate API and annotated as accepted or rejected by human annotators (we do not include rejected sentence pairs)
- task_master_1: [Taskmaster-1](https://research.google/tools/datasets/taskmaster-1/) translated by professional translators hired by [AIResearch](https://airesearch.in.th/)
- generated_reviews_translator: professional translators hired by [AIResearch](https://airesearch.in.th/)
- thai_websites: webcrawling from top 500 websites in Thailand; respective content creators; the authors only did sentence alignment
- paracrawl: replicating Paracrawl's methodology for webcrawling; respective content creators; the authors only did sentence alignment
- nus_sms: [The National University of Singapore SMS Corpus](https://scholarbank.nus.edu.sg/handle/10635/137343) translated by crowdsourced translators hired by [AIResearch](https://airesearch.in.th/)
- wikipedia: Thai Wikipedia; respective content creators; the authors only did sentence alignment
- assorted_government: Government document in PDFs from various government websites; respective content creators; the authors only did sentence alignment
- generated_reviews_crowd: generated by [CTRL](https://arxiv.org/abs/1909.05858), translated to Thai by crowdsourced translators hired by [AIResearch](https://airesearch.in.th/)
- aqdf: Bilingual news from [Asia Pacific Defense Forum](https://ipdefenseforum.com/); respective content creators; the authors only did sentence alignment
- msr_paraphrase: [Microsoft Research Paraphrase Corpus](https://www.microsoft.com/en-us/download/details.aspx?id=52398) translated to Thai by crowdsourced translators hired by [AIResearch](https://airesearch.in.th/)
- mozilla_common_voice: English version of [Mozilla Common Voice](https://commonvoice.mozilla.org/) translated to Thai by crowdsourced translators hired by [AIResearch](https://airesearch.in.th/)
### Personal and Sensitive Information
There are risks of personal information to be included in the webcrawled data namely `paracrawl` and `thai_websites`.
## Considerations for Using the Data
### Social Impact of Dataset
- The first and currently largest English-Thai machine translation dataset that is strictly cleaned and deduplicated, compare to other sources such as Paracrawl.
### Discussion of Biases
- Gender-based ending honorifics in Thai (ครับ/ค่ะ) might not be balanced due to more female translators than male for `task_master_1`
### Other Known Limitations
#### Segment Alignment between Languages With and Without Boundaries
Unlike English, there is no segment boundary marking in Thai. One segment in Thai may or may not cover all
the content of an English segment. Currently, we mitigate this problem by grouping Thai segments together before
computing the text similarity scores. We then choose the combination with the highest text similarity score. It can be
said that adequacy is the main issue in building this dataset.
Quality of Translation from Crawled Websites
Some websites use machine translation models such as Google Translate to localize their content. As a result, Thai
segments retrieved from web crawling might face issues of fluency since we do not use human annotators to perform
quality control.
#### Quality Control of Crowdsourced Translators
When we use a crowdsourcing platform to translate the content, we can not fully control the quality of the translation.
To combat this, we filter out low-quality segments by using a text similarity threshold, based on cosine similarity of
universal sentence encoder vectors. Moreover, some crowdsourced translators might copy and paste source segments to
a translation engine and take the results as answers to the platform. To further improve, we can apply techniques such
as described in [Zaidan, 2012] to control the quality and avoid fraud on the platform.
#### Domain Dependence of Machine Tranlsation Models
We test domain dependence of machine translation models by comparing models trained and tested on the same dataset,
using 80/10/10 train-validation-test split, and models trained on one dataset and tested on the other.
## Additional Information
### Dataset Curators
[AIResearch](https://airesearch.in.th/), funded by [VISTEC](https://www.vistec.ac.th/) and [depa](https://www.depa.or.th/th/home)
### Licensing Information
CC-BY-SA 4.0
### Citation Information
```
@article{lowphansirikul2020scb,
title={scb-mt-en-th-2020: A Large English-Thai Parallel Corpus},
author={Lowphansirikul, Lalita and Polpanumas, Charin and Rutherford, Attapol T and Nutanong, Sarana},
journal={arXiv preprint arXiv:2007.03541},
year={2020}
}
```
### Contributions
Thanks to [@cstorm125](https://github.com/cstorm125) for adding this dataset. |
false |
## Introduction
We propose the first question-answering dataset driven by STEM theorems. We annotated 800 QA pairs covering 350+ theorems spanning across Math, EE&CS, Physics and Finance. The dataset is collected by human experts with very high quality. We provide the dataset as a new benchmark to test the limit of large language models to apply theorems to solve challenging university-level questions. We provide a pipeline in the following to prompt LLMs and evaluate their outputs with WolframAlpha.
## How to use TheoremQA
```
from datasets import load_dataset
dataset = load_dataset("wenhu/TheoremQA")
for d in dataset['test']:
print(d)
```
To use the images, try to download images from https://github.com/wenhuchen/TheoremQA/tree/main/images. The image is under the `Picture' field.
## Arxiv Paper:
https://arxiv.org/abs/2305.12524
## Code
https://github.com/wenhuchen/TheoremQA/tree/main |
true | https://github.com/Yale-LILY/FOLIO
```
@article{han2022folio,
title={FOLIO: Natural Language Reasoning with First-Order Logic},
author = {Han, Simeng and Schoelkopf, Hailey and Zhao, Yilun and Qi, Zhenting and Riddell, Martin and Benson, Luke and Sun, Lucy and Zubova, Ekaterina and Qiao, Yujie and Burtell, Matthew and Peng, David and Fan, Jonathan and Liu, Yixin and Wong, Brian and Sailor, Malcolm and Ni, Ansong and Nan, Linyong and Kasai, Jungo and Yu, Tao and Zhang, Rui and Joty, Shafiq and Fabbri, Alexander R. and Kryscinski, Wojciech and Lin, Xi Victoria and Xiong, Caiming and Radev, Dragomir},
journal={arXiv preprint arXiv:2209.00840},
url = {https://arxiv.org/abs/2209.00840},
year={2022}
}
``` |
false |
# Dataset Card for ThaiSum
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/nakhunchumpolsathien/ThaiSum
- **Repository:** https://github.com/nakhunchumpolsathien/ThaiSum
- **Paper:**
- **Leaderboard:**
- **Point of Contact:** https://github.com/nakhunchumpolsathien
### Dataset Summary
ThaiSum is a large-scale corpus for Thai text summarization obtained from several online news websites namely Thairath, ThaiPBS, Prachathai, and The Standard. This dataset consists of over 350,000 article and summary pairs written by journalists.
### Supported Tasks and Leaderboards
summarization, language modeling
### Languages
Thai
## Dataset Structure
### Data Instances
```
{'body': 'กีเก ซานเชซ ฟลอเรส\xa0 กุนซือเลือดกระทิงของทีมวัตฟอร์ด\xa0 เมินประเด็นจุดโทษปัญหาในเกมพรีเมียร์ลีก อังกฤษ นัดที่แตนอาละวาดเปิดบ้านพ่าย คริสตัล พาเลซ 0-1ชี้ทีมของเขาเล่นไม่ดีพอเอง,สำนักข่าวต่างประเทศรายงานวันที่ 27 ก.ย. ว่า กีเก ซานเชซ ฟลอเรส\xa0 ผู้จัดการทีมชาวสเปน ของ แตนอาละวาด วัตฟอร์ด\xa0 ยอมรับทีมของเขาเล่นได้ไม่ดีพอเอง ในเกมพรีเมียร์ลีก อังกฤษ นัดเปิดบ้านพ่าย อินทรีผงาด คริสตัล พาเลซ 0-1 เมื่อคืนวันอาทิตย์ที่ผ่านมา,เกมนี้จุดเปลี่ยนมาอยู่ที่การได้จุดโทษในช่วงครึ่งหลังของ คริสตัล พาเลซ ซึ่งไม่ค่อยชัดเจนเท่าไหร่ว่า อัลลัน นียอม นั้นไปทำฟาล์วใส่ วิลฟรีด ซาฮา ในเขตโทษหรือไม่ แต่ผู้ตัดสินก็ชี้เป็นจุดโทษ ซึ่ง โยอัน กาบาย สังหารไม่พลาด และเป็นประตูชัยช่วยให้ คริสตัล พาเลซ เอาชนะ วัตฟอร์ด ไป 1-0 และเป็นการพ่ายแพ้ในบ้านนัดแรกของวัตฟอร์ดในฤดูกาลนี้อีกด้วย,ฟลอเรส กล่าวว่า มันเป็นเรื่องยากในการหยุดเกมรุกของคริสตัล พาเลซ ซึ่งมันอึดอัดจริงๆสำหรับเรา เราเล่นกันได้ไม่ดีนักในตอนที่ได้ครองบอล เราต้องเล่นทางริมเส้นให้มากกว่านี้ เราไม่สามารถหยุดเกมสวนกลับของพวกเขาได้ และแนวรับของเราก็ยืนไม่เป็นระเบียบสักเท่าไหร่ในช่วงครึ่งแรก ส่วนเรื่องจุดโทษการตัดสินใจขั้นสุดท้ายมันอยู่ที่ผู้ตัดสิน ซึ่งมันเป็นการตัดสินใจที่สำคัญ ผมเองก็ไม่รู้ว่าเขาตัดสินถูกหรือเปล่า บางทีมันอาจเป็นจุดที่ตัดสินเกมนี้เลย แต่เราไม่ได้แพ้เกมนี้เพราะจุดโทษ เราแพ้ในวันนี้เพราะเราเล่นไม่ดีและคริสตัล พาเลซ เล่นดีกว่าเรา เราไม่ได้มีฟอร์มการเล่นที่ดีในเกมนี้เลย', 'summary': 'กีเก ซานเชซ ฟลอเรส กุนซือเลือดกระทิงของทีมวัตฟอร์ด เมินประเด็นจุดโทษปัญหาในเกมพรีเมียร์ลีก อังกฤษ นัดที่แตนอาละวาดเปิดบ้านพ่าย คริสตัล พาเลซ 0-1ชี้ทีมของเขาเล่นไม่ดีพอเอง', 'tags': 'พรีเมียร์ลีก,วัตฟอร์ด,คริสตัล พาเลซ,กีเก ซานเชซ ฟลอเรส,ข่าวกีฬา,ข่าว,ไทยรัฐออนไลน์', 'title': 'ฟลอเรส รับ วัตฟอร์ดห่วยเองเกมพ่ายพาเลซคาบ้าน', 'type': '', 'url': 'https://www.thairath.co.th/content/528322'}
```
### Data Fields
- `title`: title of article
- `body`: body of article
- `summary`: summary of article
- `type`: type of article, if any
- `tags`: tags of article, separated by `,`
- `url`: URL of article
### Data Splits
train/valid/test: 358868 / 11000 / 11000
## Dataset Creation
### Curation Rationale
Sequence-to-sequence (Seq2Seq) models have shown great achievement in text summarization. However, Seq2Seq model often requires large-scale training data to achieve effective results. Although many impressive advancements in text summarization field have been made, most of summarization studies focus on resource-rich languages. The progress of Thai text summarization is still far behind. The dearth of large-scale dataset keeps Thai text summarization in its infancy. As far as our knowledge goes, there is not a large-scale dataset for Thai text summarization available anywhere. Thus, we present ThaiSum, a large-scale corpus for Thai text summarization obtained from several online news websites namely Thairath, ThaiPBS, Prachathai, and The Standard.
### Source Data
#### Initial Data Collection and Normalization
We used a python library named Scrapy to crawl articles from several news websites namely Thairath, Prachatai, ThaiPBS and, The Standard. We first collected news URLs provided in their sitemaps. During web-crawling, we used HTML markup and metadata available in HTML pages to identify article text, summary, headline, tags and label. Collected articles were published online from 2014 to August 2020. <br> <br>
We further performed data cleansing process to minimize noisy data. We filtered out articles that their article text or summary is missing. Articles that contains article text with less than 150 words or summary with less than 15 words were removed. We also discarded articles that contain at least one of these following tags: ‘ดวง’ (horoscope), ‘นิยาย’ (novel), ‘อินสตราแกรมดารา’ (celebrity Instagram), ‘คลิปสุดฮา’(funny video) and ‘สรุปข่าว’ (highlight news). Some summaries were completely irrelevant to their original article texts. To eliminate those irrelevant summaries, we calculated abstractedness score between summary and its article text. Abstractedness score is written formally as: <br>
<center><a href="https://www.codecogs.com/eqnedit.php?latex=\begin{equation}&space;\frac{|S-A|}{r}&space;\times&space;100&space;\end{equation}" target="_blank"><img src="https://latex.codecogs.com/gif.latex?\begin{equation}&space;\frac{|S-A|}{r}&space;\times&space;100&space;\end{equation}" title="\begin{equation} \frac{|S-A|}{r} \times 100 \end{equation}" /></a></center><br>
<br>Where 𝑆 denotes set of article tokens. 𝐴 denotes set of summary tokens. 𝑟 denotes a total number of summary tokens. We omitted articles that have abstractedness score at 1-grams higher than 60%.
<br><br>
It is important to point out that we used [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp), version 2.2.4, tokenizing engine = newmm, to process Thai texts in this study. It is challenging to tokenize running Thai text into words or sentences because there are not clear word/sentence delimiters in Thai language. Therefore, using different tokenization engines may result in different segment of words/sentences.
After data-cleansing process, ThaiSum dataset contains over 358,000 articles. The size of this dataset is comparable to a well-known English document summarization dataset, CNN/Dily mail dataset. Moreover, we analyse the characteristics of this dataset by measuring the abstractedness level, compassion rate, and content diversity. For more details, see [thaisum_exploration.ipynb](https://github.com/nakhunchumpolsathien/ThaiSum/blob/master/thaisum_exploration.ipynb).
#### Dataset Statistics
ThaiSum dataset consists of 358,868 articles. Average lengths of article texts and summaries are approximately 530 and 37 words respectively. As mentioned earlier, we also collected headlines, tags and labels provided in each article. Tags are similar to keywords of the article. An article normally contains several tags but a few labels. Tags can be name of places or persons that article is about while labels indicate news category (politic, entertainment, etc.). Ultimatly, ThaiSum contains 538,059 unique tags and 59 unique labels. Note that not every article contains tags or labels.
|Dataset Size| 358,868 | articles |
|:---|---:|---:|
|Avg. Article Length| 529.5 | words|
|Avg. Summary Length | 37.3 | words|
|Avg. Headline Length | 12.6 | words|
|Unique Vocabulary Size | 407,355 | words|
|Occurring > 10 times | 81,761 | words|
|Unique News Tag Size | 538,059 | tags|
|Unique News Label Size | 59 | labels|
#### Who are the source language producers?
Journalists of respective articles
### Annotations
#### Annotation process
`summary`, `type` and `tags` are created by journalists who wrote the articles and/or their publishers.
#### Who are the annotators?
`summary`, `type` and `tags` are created by journalists who wrote the articles and/or their publishers.
### Personal and Sensitive Information
All data are public news articles. No personal and sensitive information is expected to be included.
## Considerations for Using the Data
### Social Impact of Dataset
- News summarization in Thai
- Language modeling for Thai news
### Discussion of Biases
- [ThaiPBS](https://www.thaipbs.or.th/home) [receives funding from Thai government](https://www.bangkokbiznews.com/blog/detail/648740).
- [Thairath](https://www.thairath.co.th/) is known as [the most popular newspaper in Thailand](https://mgronline.com/onlinesection/detail/9620000058532); no clear political leaning.
- [The Standard](https://thestandard.co/) is a left-leaning online magazine.
- [Prachathai](https://prachatai.com/) is a left-leaning, human-right-focused news site.
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[@nakhunchumpolsathien](https://github.com/nakhunchumpolsathien/)
[@caramelWaffle](https://github.com/caramelWaffle)
### Licensing Information
MIT License
### Citation Information
```
@mastersthesis{chumpolsathien_2020,
title={Using Knowledge Distillation from Keyword Extraction to Improve the Informativeness of Neural Cross-lingual Summarization},
author={Chumpolsathien, Nakhun},
year={2020},
school={Beijing Institute of Technology}
```
### Contributions
Thanks to [@cstorm125](https://github.com/cstorm125) for adding this dataset. |
true | # Dataset Card for "code_x_glue_cc_defect_detection"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits-sample-size)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/Defect-detection
### Dataset Summary
CodeXGLUE Defect-detection dataset, available at https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/Defect-detection
Given a source code, the task is to identify whether it is an insecure code that may attack software systems, such as resource leaks, use-after-free vulnerabilities and DoS attack. We treat the task as binary classification (0/1), where 1 stands for insecure code and 0 for secure code.
The dataset we use comes from the paper Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks. We combine all projects and split 80%/10%/10% for training/dev/test.
### Supported Tasks and Leaderboards
- `multi-class-classification`: The dataset can be used to train a model for detecting if code has a defect in it.
### Languages
- C **programming** language
## Dataset Structure
### Data Instances
An example of 'validation' looks as follows.
```
{
"commit_id": "aa1530dec499f7525d2ccaa0e3a876dc8089ed1e",
"func": "static void filter_mirror_setup(NetFilterState *nf, Error **errp)\n{\n MirrorState *s = FILTER_MIRROR(nf);\n Chardev *chr;\n chr = qemu_chr_find(s->outdev);\n if (chr == NULL) {\n error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,\n \"Device '%s' not found\", s->outdev);\n qemu_chr_fe_init(&s->chr_out, chr, errp);",
"id": 8,
"project": "qemu",
"target": true
}
```
### Data Fields
In the following each data field in go is explained for each config. The data fields are the same among all splits.
#### default
|field name| type | description |
|----------|------|------------------------------------------|
|id |int32 | Index of the sample |
|func |string| The source code |
|target |bool | 0 or 1 (vulnerability or not) |
|project |string| Original project that contains this code |
|commit_id |string| Commit identifier in the original project|
### Data Splits
| name |train|validation|test|
|-------|----:|---------:|---:|
|default|21854| 2732|2732|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
https://github.com/microsoft, https://github.com/madlag
### Licensing Information
Computational Use of Data Agreement (C-UDA) License.
### Citation Information
```
@inproceedings{zhou2019devign,
title={Devign: Effective vulnerability identification by learning comprehensive program semantics via graph neural networks},
author={Zhou, Yaqin and Liu, Shangqing and Siow, Jingkai and Du, Xiaoning and Liu, Yang},
booktitle={Advances in Neural Information Processing Systems},
pages={10197--10207}, year={2019}
```
### Contributions
Thanks to @madlag (and partly also @ncoop57) for adding this dataset. |
false |
# Dataset Card for chinese_chatgpt_corpus
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Size of downloaded dataset files:** 5.05 GB
- **Size of the generated dataset:** 0 GB
- **Total amount of disk used:** 5.05 GB
### Dataset Summary
This repo collects chinese corpus for Supervised Finetuning (SFT) and Reinforcement Learning From Human Feedback (RLHF).
### Supported Tasks and Leaderboards
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Languages
Chinese
## Dataset Structure
### Data Instances
#### train_data_external_v1.jsonl
- **Size of downloaded dataset files:** 5.04 GB
- **Size of the generated dataset:** 0 GB
- **Total amount of disk used:** 5.04 GB
An example looks as follows:
```
{
"prompt": "问题：有没有给未成年贷款的有的联系",
"answers":
[
{
"answer": "若通过招行办理，我行规定，贷款人年龄需年满18岁，且年龄加贷款年限不得超过70岁。如果您持有我行信用卡附属卡，可尝试办理预借现金。",
"score": 1
}
],
"prefix": "回答："
}
```
#### dev_data_external_v1.jsonl
- **Size of downloaded dataset files:** 9.55 MB
- **Size of the generated dataset:** 0 MB
- **Total amount of disk used:** 9.55 MB
An example looks as follows:
```
{
"prompt": "初学纹发现1/2\"的管螺纹并不是1\"的一半。不知道其中的原因，请各位指点。",
"answers":
[
{
"answer": "管螺纹的名义尺寸是“管子”的孔（内）径，而管子的壁厚不是两倍。所以，1/2\"的管螺纹并不是1\"的一半，",
"score": 1
}
],
"prefix": "回答："
}
```
### Data Fields
The data fields are the same among all splits.
#### train_data_external_v1.jsonl
- `prompt`: prompt, `string`
- `answers`: list of answers
- `answer`: answer, `string`
- `score`: score of answer, `int`
- `prefix`: prefix to the answer, `string`
#### dev_data_external_v1.jsonl
- `prompt`: prompt, `string`
- `answers`: list of answers
- `answer`: answer, `string`
- `score`: score of answer, `int`
- `prefix`: prefix to the answer, `string`
### Data Splits
| name | train |
|----------|-------:|
|train_data_external_v1.jsonl|5477982|
|dev_data_external_v1.jsonl|10000|
## Dataset Creation
### Curation Rationale
Link to github: [data_prepare](https://github.com/sunzeyeah/RLHF/blob/master/src/data_prepare.py)
### Source Data
#### Initial Data Collection and Normalization
- [百科](https://github.com/brightmart/nlp_chinese_corpus)
- [知道问答](https://github.com/SophonPlus/ChineseNlpCorpus)
- [对联](https://github.com/wb14123/couplet-dataset/releases/download/1.0/couplet.tar.gz)
- [古文](https://github.com/NiuTrans/Classical-Modern)
- [古诗词](https://github.com/chinese-poetry/chinese-poetry)
- 微博新闻评论
#### Who are the source language producers?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Annotations
#### Annotation process
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
#### Who are the annotators?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Personal and Sensitive Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Discussion of Biases
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Other Known Limitations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Additional Information
### Dataset Curators
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Licensing Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
true |
# Dataset Card for the EUR-Lex dataset
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** http://nlp.cs.aueb.gr/software_and_datasets/EURLEX57K/
- **Repository:** http://nlp.cs.aueb.gr/software_and_datasets/EURLEX57K/
- **Paper:** https://www.aclweb.org/anthology/P19-1636/
- **Leaderboard:** N/A
- **Point of Contact:** [Ilias Chalkidis](mailto:ihalk@aueb.gr)
### Dataset Summary
EURLEX57K can be viewed as an improved version of the dataset released by Mencia and Furnkranzand (2007), which has been widely used in Large-scale Multi-label Text Classification (LMTC) research, but is less than half the size of EURLEX57K (19.6k documents, 4k EUROVOC labels) and more than ten years old.
EURLEX57K contains 57k legislative documents in English from EUR-Lex (https://eur-lex.europa.eu) with an average length of 727 words. Each document contains four major zones:
- the header, which includes the title and name of the legal body enforcing the legal act;
- the recitals, which are legal background references; and
- the main body, usually organized in articles.
**Labeling / Annotation**
All the documents of the dataset have been annotated by the Publications Office of EU (https://publications.europa.eu/en) with multiple concepts from EUROVOC (http://eurovoc.europa.eu/).
While EUROVOC includes approx. 7k concepts (labels), only 4,271 (59.31%) are present in EURLEX57K, from which only 2,049 (47.97%) have been assigned to more than 10 documents. The 4,271 labels are also divided into frequent (746 labels), few-shot (3,362), and zero- shot (163), depending on whether they were assigned to more than 50, fewer than 50 but at least one, or no training documents, respectively.
### Supported Tasks and Leaderboards
The dataset supports:
**Multi-label Text Classification:** Given the text of a document, a model predicts the relevant EUROVOC concepts.
**Few-shot and Zero-shot learning:** As already noted, the labels can be divided into three groups: frequent (746 labels), few-shot (3,362), and zero- shot (163), depending on whether they were assigned to more than 50, fewer than 50 but at least one, or no training documents, respectively.
### Languages
All documents are written in English.
## Dataset Structure
### Data Instances
```json
{
"celex_id": "31979D0509",
"title": "79/509/EEC: Council Decision of 24 May 1979 on financial aid from the Community for the eradication of African swine fever in Spain",
"text": "COUNCIL DECISION of 24 May 1979 on financial aid from the Community for the eradication of African swine fever in Spain (79/509/EEC)\nTHE COUNCIL OF THE EUROPEAN COMMUNITIES\nHaving regard to the Treaty establishing the European Economic Community, and in particular Article 43 thereof,\nHaving regard to the proposal from the Commission (1),\nHaving regard to the opinion of the European Parliament (2),\nWhereas the Community should take all appropriate measures to protect itself against the appearance of African swine fever on its territory;\nWhereas to this end the Community has undertaken, and continues to undertake, action designed to contain outbreaks of this type of disease far from its frontiers by helping countries affected to reinforce their preventive measures ; whereas for this purpose Community subsidies have already been granted to Spain;\nWhereas these measures have unquestionably made an effective contribution to the protection of Community livestock, especially through the creation and maintenance of a buffer zone north of the river Ebro;\nWhereas, however, in the opinion of the Spanish authorities themselves, the measures so far implemented must be reinforced if the fundamental objective of eradicating the disease from the entire country is to be achieved;\nWhereas the Spanish authorities have asked the Community to contribute to the expenses necessary for the efficient implementation of a total eradication programme;\nWhereas a favourable response should be given to this request by granting aid to Spain, having regard to the undertaking given by that country to protect the Community against African swine fever and to eliminate completely this disease by the end of a five-year eradication plan;\nWhereas this eradication plan must include certain measures which guarantee the effectiveness of the action taken, and it must be possible to adapt these measures to developments in the situation by means of a procedure establishing close cooperation between the Member States and the Commission;\nWhereas it is necessary to keep the Member States regularly informed as to the progress of the action undertaken,",
"eurovoc_concepts": ["192", "2356", "2560", "862", "863"]
}
```
### Data Fields
The following data fields are provided for documents (`train`, `dev`, `test`):
`celex_id`: (**str**) The official ID of the document. The CELEX number is the unique identifier for all publications in both Eur-Lex and CELLAR.\
`title`: (**str**) The title of the document.\
`text`: (**str**) The full content of each document, which is represented by its `header`, `recitals` and `main_body`.\
`eurovoc_concepts`: (**List[str]**) The relevant EUROVOC concepts (labels).
If you want to use the descriptors of EUROVOC concepts, similar to Chalkidis et al. (2020), please load: https://archive.org/download/EURLEX57K/eurovoc_concepts.jsonl
```python
import json
with open('./eurovoc_concepts.jsonl') as jsonl_file:
eurovoc_concepts = {json.loads(concept) for concept in jsonl_file.readlines()}
```
### Data Splits
| Split | No of Documents | Avg. words | Avg. labels |
| ------------------- | ------------------------------------ | --- | --- |
| Train | 45,000 | 729 | 5 |
|Development | 6,000 | 714 | 5 |
|Test | 6,000 | 725 | 5 |
## Dataset Creation
### Curation Rationale
The dataset was curated by Chalkidis et al. (2019).\
The documents have been annotated by the Publications Office of EU (https://publications.europa.eu/en).
### Source Data
#### Initial Data Collection and Normalization
The original data are available at EUR-Lex portal (https://eur-lex.europa.eu) in an unprocessed format.
The documents were downloaded from EUR-Lex portal in HTML format.
The relevant metadata and EUROVOC concepts were downloaded from the SPARQL endpoint of the Publications Office of EU (http://publications.europa.eu/webapi/rdf/sparql).
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
* The original documents are available at EUR-Lex portal (https://eur-lex.europa.eu) in an unprocessed HTML format. The HTML code was striped and the documents split into sections.
* The documents have been annotated by the Publications Office of EU (https://publications.europa.eu/en).
#### Who are the annotators?
Publications Office of EU (https://publications.europa.eu/en)
### Personal and Sensitive Information
The dataset does not include personal or sensitive information.
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Chalkidis et al. (2019)
### Licensing Information
© European Union, 1998-2021
The Commission’s document reuse policy is based on Decision 2011/833/EU. Unless otherwise specified, you can re-use the legal documents published in EUR-Lex for commercial or non-commercial purposes.
The copyright for the editorial content of this website, the summaries of EU legislation and the consolidated texts, which is owned by the EU, is licensed under the Creative Commons Attribution 4.0 International licence. This means that you can re-use the content provided you acknowledge the source and indicate any changes you have made.
Source: https://eur-lex.europa.eu/content/legal-notice/legal-notice.html \
Read more: https://eur-lex.europa.eu/content/help/faq/reuse-contents-eurlex.html
### Citation Information
*Ilias Chalkidis, Manos Fergadiotis, Prodromos Malakasiotis and Ion Androutsopoulos.*
*Large-Scale Multi-Label Text Classification on EU Legislation.*
*Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019). Florence, Italy. 2019*
```
@inproceedings{chalkidis-etal-2019-large,
title = "Large-Scale Multi-Label Text Classification on {EU} Legislation",
author = "Chalkidis, Ilias and Fergadiotis, Manos and Malakasiotis, Prodromos and Androutsopoulos, Ion",
booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
year = "2019",
address = "Florence, Italy",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/P19-1636",
doi = "10.18653/v1/P19-1636",
pages = "6314--6322"
}
```
### Contributions
Thanks to [@iliaschalkidis](https://github.com/iliaschalkidis) for adding this dataset. |
false |
# Dataset Card for CORD-19
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://www.semanticscholar.org/cord19](https://www.semanticscholar.org/cord19)
- **Repository:** [https://github.com/allenai/cord19](https://github.com/allenai/cord19)
- **Paper:** [CORD-19: The COVID-19 Open Research Dataset](https://www.aclweb.org/anthology/2020.nlpcovid19-acl.1/)
- **Leaderboard:** [Kaggle challenge](https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-challenge)
### Dataset Summary
CORD-19 is a corpus of academic papers about COVID-19 and related coronavirus research. It's curated and maintained by the Semantic Scholar team at the Allen Institute for AI to support text mining and NLP research.
### Supported Tasks and Leaderboards
See tasks defined in the related [Kaggle challenge](https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-challenge/tasks)
### Languages
The dataset is in english (en).
## Dataset Structure
### Data Instances
The following code block present an overview of a sample in json-like syntax (abbreviated since some fields are very long):
```
{
"abstract": "OBJECTIVE: This retrospective chart review describes the epidemiology and clinical features of 40 patients with culture-proven Mycoplasma pneumoniae infections at King Abdulaziz University Hospital, Jeddah, Saudi Arabia. METHODS: Patients with positive M. pneumoniae cultures from respiratory specimens from January 1997 through December 1998 were identified through the Microbiology records. Charts of patients were reviewed. RESULTS: 40 patients were identified [...]",
"authors": "Madani, Tariq A; Al-Ghamdi, Aisha A",
"cord_uid": "ug7v899j",
"doc_embeddings": [
-2.939983606338501,
-6.312200546264648,
-1.0459030866622925,
[...] 766 values in total [...]
-4.107113361358643,
-3.8174145221710205,
1.8976187705993652,
5.811529159545898,
-2.9323840141296387
],
"doi": "10.1186/1471-2334-1-6",
"journal": "BMC Infect Dis",
"publish_time": "2001-07-04",
"sha": "d1aafb70c066a2068b02786f8929fd9c900897fb",
"source_x": "PMC",
"title": "Clinical features of culture-proven Mycoplasma pneumoniae infections at King Abdulaziz University Hospital, Jeddah, Saudi Arabia",
"url": "https: //www.ncbi.nlm.nih.gov/pmc/articles/PMC35282/"
}
```
### Data Fields
Currently only the following fields are integrated: `cord_uid`, `sha`,`source_x`, `title`, `doi`, `abstract`, `publish_time`, `authors`, `journal`. With `fulltext` configuration, the sections transcribed in `pdf_json_files` are converted in `fulltext` feature.
- `cord_uid`: A `str`-valued field that assigns a unique identifier to each CORD-19 paper. This is not necessariy unique per row, which is explained in the FAQs.
- `sha`: A `List[str]`-valued field that is the SHA1 of all PDFs associated with the CORD-19 paper. Most papers will have either zero or one value here (since we either have a PDF or we don't), but some papers will have multiple. For example, the main paper might have supplemental information saved in a separate PDF. Or we might have two separate PDF copies of the same paper. If multiple PDFs exist, their SHA1 will be semicolon-separated (e.g. `'4eb6e165ee705e2ae2a24ed2d4e67da42831ff4a; d4f0247db5e916c20eae3f6d772e8572eb828236'`)
- `source_x`: A `List[str]`-valued field that is the names of sources from which we received this paper. Also semicolon-separated. For example, `'ArXiv; Elsevier; PMC; WHO'`. There should always be at least one source listed.
- `title`: A `str`-valued field for the paper title
- `doi`: A `str`-valued field for the paper DOI
- `pmcid`: A `str`-valued field for the paper's ID on PubMed Central. Should begin with `PMC` followed by an integer.
- `pubmed_id`: An `int`-valued field for the paper's ID on PubMed.
- `license`: A `str`-valued field with the most permissive license we've found associated with this paper. Possible values include: `'cc0', 'hybrid-oa', 'els-covid', 'no-cc', 'cc-by-nc-sa', 'cc-by', 'gold-oa', 'biorxiv', 'green-oa', 'bronze-oa', 'cc-by-nc', 'medrxiv', 'cc-by-nd', 'arxiv', 'unk', 'cc-by-sa', 'cc-by-nc-nd'`
- `abstract`: A `str`-valued field for the paper's abstract
- `publish_time`: A `str`-valued field for the published date of the paper. This is in `yyyy-mm-dd` format. Not always accurate as some publishers will denote unknown dates with future dates like `yyyy-12-31`
- `authors`: A `List[str]`-valued field for the authors of the paper. Each author name is in `Last, First Middle` format and semicolon-separated.
- `journal`: A `str`-valued field for the paper journal. Strings are not normalized (e.g. `BMJ` and `British Medical Journal` can both exist). Empty string if unknown.
- `mag_id`: Deprecated, but originally an `int`-valued field for the paper as represented in the Microsoft Academic Graph.
- `who_covidence_id`: A `str`-valued field for the ID assigned by the WHO for this paper. Format looks like `#72306`.
- `arxiv_id`: A `str`-valued field for the arXiv ID of this paper.
- `pdf_json_files`: A `List[str]`-valued field containing paths from the root of the current data dump version to the parses of the paper PDFs into JSON format. Multiple paths are semicolon-separated. Example: `document_parses/pdf_json/4eb6e165ee705e2ae2a24ed2d4e67da42831ff4a.json; document_parses/pdf_json/d4f0247db5e916c20eae3f6d772e8572eb828236.json`
- `pmc_json_files`: A `List[str]`-valued field. Same as above, but corresponding to the full text XML files downloaded from PMC, parsed into the same JSON format as above.
- `url`: A `List[str]`-valued field containing all URLs associated with this paper. Semicolon-separated.
- `s2_id`: A `str`-valued field containing the Semantic Scholar ID for this paper. Can be used with the Semantic Scholar API (e.g. `s2_id=9445722` corresponds to `http://api.semanticscholar.org/corpusid:9445722`)
Extra fields based on selected configuration during loading:
- `fulltext`: A `str`-valued field containing the concatenation of all text sections from json (itself extracted from pdf)
- `doc_embeddings`: A `sequence` of float-valued elements containing document embeddings as a vector of floats (parsed from string of values separated by ','). Details on the system used to extract the embeddings are available in: [SPECTER: Document-level Representation Learning using Citation-informed Transformers](https://arxiv.org/abs/2004.07180). TL;DR: it's relying on a BERT model pre-trained on document-level relatedness using the citation graph. The system can be queried through REST (see [public API documentation](https://github.com/allenai/paper-embedding-public-apis)).
### Data Splits
No annotation provided in this dataset so all instances are provided in training split.
The sizes of each configuration are:
| | train |
|------------|-------:|
| metadata | 368618 |
| fulltext | 368618 |
| embeddings | 368618 |
## Dataset Creation
### Curation Rationale
See [official readme](https://github.com/allenai/cord19/blob/master/README.md)
### Source Data
See [official readme](https://github.com/allenai/cord19/blob/master/README.md)
#### Initial Data Collection and Normalization
See [official readme](https://github.com/allenai/cord19/blob/master/README.md)
#### Who are the source language producers?
See [official readme](https://github.com/allenai/cord19/blob/master/README.md)
### Annotations
No annotations
#### Annotation process
N/A
#### Who are the annotators?
N/A
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@article{Wang2020CORD19TC,
title={CORD-19: The Covid-19 Open Research Dataset},
author={Lucy Lu Wang and Kyle Lo and Yoganand Chandrasekhar and Russell Reas and Jiangjiang Yang and Darrin Eide and
K. Funk and Rodney Michael Kinney and Ziyang Liu and W. Merrill and P. Mooney and D. Murdick and Devvret Rishi and
Jerry Sheehan and Zhihong Shen and B. Stilson and A. Wade and K. Wang and Christopher Wilhelm and Boya Xie and
D. Raymond and Daniel S. Weld and Oren Etzioni and Sebastian Kohlmeier},
journal={ArXiv},
year={2020}
}
```
### Contributions
Thanks to [@ggdupont](https://github.com/ggdupont) for adding this dataset. |
false |
# Dataset Card for Multi-LexSum
## Table of Contents
- [Dataset Card for Multi-LexSum](#dataset-card-for-multi-lexsum)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Languages](#languages)
- [Dataset](#dataset)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Sheet (Datasheet)](#dataset-sheet-datasheet)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Release History](#release-history)
## Dataset Description
- **Homepage:** https://multilexsum.github.io
- **Repository:** https://github.com/multilexsum/dataset
- **Paper:** https://arxiv.org/abs/2206.10883
<p>
<a href="https://multilexsum.github.io" style="display: inline-block;">
<img src="https://img.shields.io/badge/-homepage-informational.svg?logo=jekyll" title="Multi-LexSum Paper" style="margin-top: 0.25rem; margin-bottom: 0.25rem"></a>
<a href="https://github.com/multilexsum/dataset" style="display: inline-block;">
<img src="https://img.shields.io/badge/-multilexsum-lightgrey.svg?logo=github" title="Multi-LexSum Github Repo" style="margin-top: 0.25rem; margin-bottom: 0.25rem"></a>
<a href="https://arxiv.org/abs/2206.10883" style="display: inline-block;">
<img src="https://img.shields.io/badge/NeurIPS-2022-9cf" title="Multi-LexSum is accepted in NeurIPS 2022" style="margin-top: 0.25rem; margin-bottom: 0.25rem"></a>
</p>
### Talk @ NeurIPS 2022
[](https://youtu.be/C-fwW_ZhkE8)
### Dataset Summary
The Multi-LexSum dataset is a collection of 9,280 such legal case summaries. Multi-LexSum is distinct from other datasets in its **multiple target summaries, each at a different granularity** (ranging from one-sentence “extreme” summaries to multi-paragraph narrations of over five hundred words). It presents a challenging multi-document summarization task given **the long length of the source documents**, often exceeding two hundred pages per case. Unlike other summarization datasets that are (semi-)automatically curated, Multi-LexSum consists of **expert-authored summaries**: the experts—lawyers and law students—are trained to follow carefully created guidelines, and their work is reviewed by an additional expert to ensure quality.
### Languages
English
## Dataset
### Data Fields
The dataset contains a list of instances (cases); each instance contains the following data:
| Field | Description |
| ------------: | -------------------------------------------------------------------------------: |
| id | `(str)` The case ID |
| sources | `(List[str])` A list of strings for the text extracted from the source documents |
| summary/long | `(str)` The long (multi-paragraph) summary for this case |
| summary/short | `(Optional[str])` The short (one-paragraph) summary for this case |
| summary/tiny | `(Optional[str])` The tiny (one-sentence) summary for this case |
Please check the exemplar usage below for loading the data:
```python
from datasets import load_dataset
multi_lexsum = load_dataset("allenai/multi_lexsum", name="v20230518")
# Download multi_lexsum locally and load it as a Dataset object
example = multi_lexsum["validation"][0] # The first instance of the dev set
example["sources"] # A list of source document text for the case
for sum_len in ["long", "short", "tiny"]:
print(example["summary/" + sum_len]) # Summaries of three lengths
print(example['case_metadata']) # The corresponding metadata for a case in a dict
```
### Data Splits
| | Instances | Source Documents (D) | Long Summaries (L) | Short Summaries (S) | Tiny Summaries (T) | Total Summaries |
| ----------: | --------: | -------------------: | -----------------: | ------------------: | -----------------: | --------------: |
| Train (70%) | 3,177 | 28,557 | 3,177 | 2,210 | 1,130 | 6,517 |
| Test (20%) | 908 | 7,428 | 908 | 616 | 312 | 1,836 |
| Dev (10%) | 454 | 4,134 | 454 | 312 | 161 | 927 |
## Dataset Sheet (Datasheet)
Please check our [dataset sheet](https://multilexsum.github.io/datasheet) for details regarding dataset creation, source data, annotation, and considerations for the usage.
## Additional Information
### Dataset Curators
The dataset is created by the collaboration between Civil Rights Litigation Clearinghouse (CRLC, from University of Michigan) and Allen Institute for AI. Multi-LexSum builds on the dataset used and posted by the Clearinghouse to inform the public about civil rights litigation.
### Licensing Information
The Multi-LexSum dataset is distributed under the [Open Data Commons Attribution License (ODC-By)](https://opendatacommons.org/licenses/by/1-0/).
The case summaries and metadata are licensed under the [Creative Commons Attribution License (CC BY-NC)](https://creativecommons.org/licenses/by-nc/4.0/), and the source documents are already in the public domain.
Commercial users who desire a license for summaries and metadata can contact [info@clearinghouse.net](mailto:info@clearinghouse.net), which will allow free use but limit summary re-posting.
The corresponding code for downloading and loading the dataset is licensed under the Apache License 2.0.
### Citation Information
```
@article{Shen2022MultiLexSum,
author = {Zejiang Shen and
Kyle Lo and
Lauren Yu and
Nathan Dahlberg and
Margo Schlanger and
Doug Downey},
title = {Multi-LexSum: Real-World Summaries of Civil Rights Lawsuits at Multiple Granularities},
journal = {CoRR},
volume = {abs/2206.10883},
year = {2022},****
url = {https://doi.org/10.48550/arXiv.2206.10883},
doi = {10.48550/arXiv.2206.10883}
}
```
## Release History
| Version | Description |
| ----------: | -----------------------------------------------------------: |
| `v20230518` | The v1.1 release including case and source document metadata |
| `v20220616` | The initial v1.0 release | |
true |
# Dataset Card for Offenseval Dravidian
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://competitions.codalab.org/competitions/27654#learn_the_details
- **Repository:** https://competitions.codalab.org/competitions/27654#participate-get_data
- **Paper:** Findings of the Shared Task on {O}ffensive {L}anguage {I}dentification in {T}amil, {M}alayalam, and {K}annada
- **Leaderboard:** https://competitions.codalab.org/competitions/27654#results
- **Point of Contact:** [Bharathi Raja Chakravarthi](mailto:bharathiraja.akr@gmail.com)
### Dataset Summary
Offensive language identification is classification task in natural language processing (NLP) where the aim is to moderate and minimise offensive content in social media. It has been an active area of research in both academia and industry for the past two decades. There is an increasing demand for offensive language identification on social media texts which are largely code-mixed. Code-mixing is a prevalent phenomenon in a multilingual community and the code-mixed texts are sometimes written in non-native scripts. Systems trained on monolingual data fail on code-mixed data due to the complexity of code-switching at different linguistic levels in the text. This shared task presents a new gold standard corpus for offensive language identification of code-mixed text in Dravidian languages (Tamil-English, Malayalam-English, and Kannada-English).
### Supported Tasks and Leaderboards
The goal of this task is to identify offensive language content of the code-mixed dataset of comments/posts in Dravidian Languages ( (Tamil-English, Malayalam-English, and Kannada-English)) collected from social media. The comment/post may contain more than one sentence but the average sentence length of the corpora is 1. Each comment/post is annotated at the comment/post level. This dataset also has class imbalance problems depicting real-world scenarios.
### Languages
Code-mixed text in Dravidian languages (Tamil-English, Malayalam-English, and Kannada-English).
## Dataset Structure
### Data Instances
An example from the Tamil dataset looks as follows:
| text | label |
| :------ | :----- |
| படம் கண்டிப்பாக வெற்றி பெற வேண்டும் செம்ம vara level | Not_offensive |
| Avasara patutiya editor uhh antha bullet sequence aa nee soliruka kudathu, athu sollama iruntha movie ku konjam support aa surprise element aa irunthurukum | Not_offensive |
An example from the Malayalam dataset looks as follows:
| text | label |
| :------ | :----- |
| ഷൈലോക്ക് ന്റെ നല്ല ടീസർ ആയിട്ട് പോലും ട്രോളി നടന്ന ലാലേട്ടൻ ഫാൻസിന് കിട്ടിയൊരു നല്ലൊരു തിരിച്ചടി തന്നെ ആയിരിന്നു ബിഗ് ബ്രദർ ന്റെ ട്രെയ്ലർ | Not_offensive |
| Marana mass Ekka kku kodukku oru | Not_offensive |
An example from the Kannada dataset looks as follows:
| text | label |
| :------ | :----- |
| ನಿಜವಾಗಿಯೂ ಅದ್ಭುತ heartly heltidini... plz avrigella namma nimmellara supprt beku | Not_offensive |
| Next song gu kuda alru andre evaga yar comment madidera alla alrru like madi share madi nam industry na next level ge togond hogaona. | Not_offensive |
### Data Fields
Tamil
- `text`: Tamil-English code mixed comment.
- `label`: integer from 0 to 5 that corresponds to these values: "Not_offensive", "Offensive_Untargetede", "Offensive_Targeted_Insult_Individual", "Offensive_Targeted_Insult_Group", "Offensive_Targeted_Insult_Other", "not-Tamil"
Malayalam
- `text`: Malayalam-English code mixed comment.
- `label`: integer from 0 to 5 that corresponds to these values: "Not_offensive", "Offensive_Untargetede", "Offensive_Targeted_Insult_Individual", "Offensive_Targeted_Insult_Group", "Offensive_Targeted_Insult_Other", "not-malayalam"
Kannada
- `text`: Kannada-English code mixed comment.
- `label`: integer from 0 to 5 that corresponds to these values: "Not_offensive", "Offensive_Untargetede", "Offensive_Targeted_Insult_Individual", "Offensive_Targeted_Insult_Group", "Offensive_Targeted_Insult_Other", "not-Kannada"
### Data Splits
| | train | validation |
|-----------|------:|-----------:|
| Tamil | 35139 | 4388 |
| Malayalam | 16010 | 1999 |
| Kannada | 6217 | 777 |
## Dataset Creation
### Curation Rationale
There is an increasing demand for offensive language identification on social media texts which are largely code-mixed. Code-mixing is a prevalent phenomenon in a multilingual community and the code-mixed texts are sometimes written in non-native scripts. Systems trained on monolingual data fail on code-mixed data due to the complexity of code-switching at different linguistic levels in the text.
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
Youtube users
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
This work is licensed under a [Creative Commons Attribution 4.0 International Licence](http://creativecommons.org/licenses/by/4.0/.)
### Citation Information
```
@article{chakravarthi-etal-2021-lre,
title = "DravidianCodeMix: Sentiment Analysis and Offensive Language Identification Dataset for Dravidian Languages in Code-Mixed Text",
author = "Chakravarthi, Bharathi Raja and
Priyadharshini, Ruba and
Muralidaran, Vigneshwaran and
Jose, Navya and
Suryawanshi, Shardul and
Sherly, Elizabeth and
McCrae, John P",
journal={Language Resources and Evaluation},
publisher={Springer}
}
```
```
@inproceedings{dravidianoffensive-eacl,
title={Findings of the Shared Task on {O}ffensive {L}anguage {I}dentification in {T}amil, {M}alayalam, and {K}annada},
author={Chakravarthi, Bharathi Raja and
Priyadharshini, Ruba and
Jose, Navya and
M, Anand Kumar and
Mandl, Thomas and
Kumaresan, Prasanna Kumar and
Ponnsamy, Rahul and
V,Hariharan and
Sherly, Elizabeth and
McCrae, John Philip },
booktitle = "Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages",
month = April,
year = "2021",
publisher = "Association for Computational Linguistics",
year={2021}
}
```
```
@inproceedings{hande-etal-2020-kancmd,
title = "{K}an{CMD}: {K}annada {C}ode{M}ixed Dataset for Sentiment Analysis and Offensive Language Detection",
author = "Hande, Adeep and
Priyadharshini, Ruba and
Chakravarthi, Bharathi Raja",
booktitle = "Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.peoples-1.6",
pages = "54--63",
abstract = "We introduce Kannada CodeMixed Dataset (KanCMD), a multi-task learning dataset for sentiment analysis and offensive language identification. The KanCMD dataset highlights two real-world issues from the social media text. First, it contains actual comments in code mixed text posted by users on YouTube social media, rather than in monolingual text from the textbook. Second, it has been annotated for two tasks, namely sentiment analysis and offensive language detection for under-resourced Kannada language. Hence, KanCMD is meant to stimulate research in under-resourced Kannada language on real-world code-mixed social media text and multi-task learning. KanCMD was obtained by crawling the YouTube, and a minimum of three annotators annotates each comment. We release KanCMD 7,671 comments for multitask learning research purpose.",
}
```
```
@inproceedings{chakravarthi-etal-2020-corpus,
title = "Corpus Creation for Sentiment Analysis in Code-Mixed {T}amil-{E}nglish Text",
author = "Chakravarthi, Bharathi Raja and
Muralidaran, Vigneshwaran and
Priyadharshini, Ruba and
McCrae, John Philip",
booktitle = "Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)",
month = may,
year = "2020",
address = "Marseille, France",
publisher = "European Language Resources association",
url = "https://www.aclweb.org/anthology/2020.sltu-1.28",
pages = "202--210",
abstract = "Understanding the sentiment of a comment from a video or an image is an essential task in many applications. Sentiment analysis of a text can be useful for various decision-making processes. One such application is to analyse the popular sentiments of videos on social media based on viewer comments. However, comments from social media do not follow strict rules of grammar, and they contain mixing of more than one language, often written in non-native scripts. Non-availability of annotated code-mixed data for a low-resourced language like Tamil also adds difficulty to this problem. To overcome this, we created a gold standard Tamil-English code-switched, sentiment-annotated corpus containing 15,744 comment posts from YouTube. In this paper, we describe the process of creating the corpus and assigning polarities. We present inter-annotator agreement and show the results of sentiment analysis trained on this corpus as a benchmark.",
language = "English",
ISBN = "979-10-95546-35-1",
}
```
```
@inproceedings{chakravarthi-etal-2020-sentiment,
title = "A Sentiment Analysis Dataset for Code-Mixed {M}alayalam-{E}nglish",
author = "Chakravarthi, Bharathi Raja and
Jose, Navya and
Suryawanshi, Shardul and
Sherly, Elizabeth and
McCrae, John Philip",
booktitle = "Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)",
month = may,
year = "2020",
address = "Marseille, France",
publisher = "European Language Resources association",
url = "https://www.aclweb.org/anthology/2020.sltu-1.25",
pages = "177--184",
abstract = "There is an increasing demand for sentiment analysis of text from social media which are mostly code-mixed. Systems trained on monolingual data fail for code-mixed data due to the complexity of mixing at different levels of the text. However, very few resources are available for code-mixed data to create models specific for this data. Although much research in multilingual and cross-lingual sentiment analysis has used semi-supervised or unsupervised methods, supervised methods still performs better. Only a few datasets for popular languages such as English-Spanish, English-Hindi, and English-Chinese are available. There are no resources available for Malayalam-English code-mixed data. This paper presents a new gold standard corpus for sentiment analysis of code-mixed text in Malayalam-English annotated by voluntary annotators. This gold standard corpus obtained a Krippendorff{'}s alpha above 0.8 for the dataset. We use this new corpus to provide the benchmark for sentiment analysis in Malayalam-English code-mixed texts.",
language = "English",
ISBN = "979-10-95546-35-1",
}
```
### Contributions
Thanks to [@jamespaultg](https://github.com/jamespaultg) for adding this dataset. |
true |
## Dataset Description
- **Homepage:** http://hatespeech.berkeley.edu
- **Paper:** https://arxiv.org/abs/2009.10277
# Dataset card for _Measuring Hate Speech_
This is a public release of the dataset described in Kennedy et al. (2020) and Sachdeva et al. (2022), consisting of 39,565 comments annotated by 7,912 annotators, for 135,556 combined rows. The primary outcome variable is the "hate speech score" but the 10 constituent ordinal labels (sentiment, (dis)respect, insult, humiliation, inferior status, violence, dehumanization, genocide, attack/defense, hate speech benchmark) can also be treated as outcomes. Includes 8 target identity groups (race/ethnicity, religion, national origin/citizenship, gender, sexual orientation, age, disability, political ideology) and 42 target identity subgroups, as well as 6 annotator demographics and 40 subgroups. The hate speech score incorporates an IRT adjustment by estimating variation in annotator interpretation of the labeling guidelines.
This dataset card is a work in progress and will be improved over time.
## Key dataset columns
* hate_speech_score - continuous hate speech measure, where higher = more hateful and lower = less hateful. > 0.5 is approximately hate speech, < -1 is counter or supportive speech, and -1 to +0.5 is neutral or ambiguous.
* text - lightly processed text of a social media post
* comment\_id - unique ID for each comment
* annotator\_id - unique ID for each annotator
* sentiment - ordinal label that is combined into the continuous score
* respect - ordinal label that is combined into the continuous score
* insult - ordinal label that is combined into the continuous score
* humiliate - ordinal label that is combined into the continuous score
* status - ordinal label that is combined into the continuous score
* dehumanize - ordinal label that is combined into the continuous score
* violence - ordinal label that is combined into the continuous score
* genocide - ordinal label that is combined into the continuous score
* attack\_defend - ordinal label that is combined into the continuous score
* hatespeech - ordinal label that is combined into the continuous score
* annotator_severity - annotator's estimated survey interpretation bias
## Code to download
The dataset can be downloaded using the following python code:
```python
import datasets
dataset = datasets.load_dataset('ucberkeley-dlab/measuring-hate-speech', 'binary')
df = dataset['train'].to_pandas()
df.describe()
```
## Citation
```
@article{kennedy2020constructing,
title={Constructing interval variables via faceted Rasch measurement and multitask deep learning: a hate speech application},
author={Kennedy, Chris J and Bacon, Geoff and Sahn, Alexander and von Vacano, Claudia},
journal={arXiv preprint arXiv:2009.10277},
year={2020}
}
```
## Contributions
Dataset curated by [@ck37](https://github.com/ck37), [@pssachdeva](https://github.com/pssachdeva), et al.
## References
Kennedy, C. J., Bacon, G., Sahn, A., & von Vacano, C. (2020). [Constructing interval variables via faceted Rasch measurement and multitask deep learning: a hate speech application](https://arxiv.org/abs/2009.10277). arXiv preprint arXiv:2009.10277.
Pratik Sachdeva, Renata Barreto, Geoff Bacon, Alexander Sahn, Claudia von Vacano, and Chris Kennedy. 2022. [The Measuring Hate Speech Corpus: Leveraging Rasch Measurement Theory for Data Perspectivism](https://aclanthology.org/2022.nlperspectives-1.11/). In *Proceedings of the 1st Workshop on Perspectivist Approaches to NLP @LREC2022*, pages 83–94, Marseille, France. European Language Resources Association. |
false | # Dataset Card for C3
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** []()
- **Repository:** [link]()
- **Paper:** []()
- **Leaderboard:** []()
- **Point of Contact:** []()
### Dataset Summary
Machine reading comprehension tasks require a machine reader to answer questions relevant to the given document. In this paper, we present the first free-form multiple-Choice Chinese machine reading Comprehension dataset (C^3), containing 13,369 documents (dialogues or more formally written mixed-genre texts) and their associated 19,577 multiple-choice free-form questions collected from Chinese-as-a-second-language examinations.
We present a comprehensive analysis of the prior knowledge (i.e., linguistic, domain-specific, and general world knowledge) needed for these real-world problems. We implement rule-based and popular neural methods and find that there is still a significant performance gap between the best performing model (68.5%) and human readers (96.0%), especially on problems that require prior knowledge. We further study the effects of distractor plausibility and data augmentation based on translated relevant datasets for English on model performance. We expect C^3 to present great challenges to existing systems as answering 86.8% of questions requires both knowledge within and beyond the accompanying document, and we hope that C^3 can serve as a platform to study how to leverage various kinds of prior knowledge to better understand a given written or orally oriented text.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
[More Information Needed]
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
[More Information Needed]
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
[More Information Needed]
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
Dataset provided for research purposes only. Please check dataset license for additional information.
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@article{sun2019investigating,
title={Investigating Prior Knowledge for Challenging Chinese Machine Reading Comprehension},
author={Sun, Kai and Yu, Dian and Yu, Dong and Cardie, Claire},
journal={Transactions of the Association for Computational Linguistics},
year={2020},
url={https://arxiv.org/abs/1904.09679v3}
}
```
### Contributions
Thanks to [@Narsil](https://github.com/Narsil) for adding this dataset. |
false |
# Dataset Card for WikiDoc
For the dataset containing rephrased content from the living textbook refer to [this dataset](https://huggingface.co/datasets/medalpaca/medical_meadow_wikidoc)
## Dataset Description
- **Source:** https://www.wikidoc.org/index.php/Main_Page
- **Repository:** https://github.com/kbressem/medalpaca
- **Paper:** TBA
### Dataset Summary
This dataset containes medical question-answer pairs extracted from [WikiDoc](https://www.wikidoc.org/index.php/Main_Page),
a collaborative platform for medical professionals to share and contribute to up-to-date medical knowledge.
The platform has to main subsites, the "Living Textbook" and "Patient Information". The "Living Textbook"
contains chapters for various medical specialties, which we crawled. We then used GTP-3.5-Turbo to rephrase
the paragraph heading to a question and used the paragraph as answer. Patient Information is structured differently,
in that each section subheading is already a question, making rephrasing them obsolete.
**Note:** This dataset is still a WIP. While the Q/A pairs from the patient information seems to be mostly correct,
the conversion using GPT-3.5-Turbo yielded some unsatisfactory results in approximately 30% of cases. We are in the process of cleaning this dataset.
### Citation Information
TBA |
false |
<div align="center">
<img width="640" alt="keremberke/pokemon-classification" src="https://huggingface.co/datasets/keremberke/pokemon-classification/resolve/main/thumbnail.jpg">
</div>
### Dataset Labels
```
['Porygon', 'Goldeen', 'Hitmonlee', 'Hitmonchan', 'Gloom', 'Aerodactyl', 'Mankey', 'Seadra', 'Gengar', 'Venonat', 'Articuno', 'Seaking', 'Dugtrio', 'Machop', 'Jynx', 'Oddish', 'Dodrio', 'Dragonair', 'Weedle', 'Golduck', 'Flareon', 'Krabby', 'Parasect', 'Ninetales', 'Nidoqueen', 'Kabutops', 'Drowzee', 'Caterpie', 'Jigglypuff', 'Machamp', 'Clefairy', 'Kangaskhan', 'Dragonite', 'Weepinbell', 'Fearow', 'Bellsprout', 'Grimer', 'Nidorina', 'Staryu', 'Horsea', 'Electabuzz', 'Dratini', 'Machoke', 'Magnemite', 'Squirtle', 'Gyarados', 'Pidgeot', 'Bulbasaur', 'Nidoking', 'Golem', 'Dewgong', 'Moltres', 'Zapdos', 'Poliwrath', 'Vulpix', 'Beedrill', 'Charmander', 'Abra', 'Zubat', 'Golbat', 'Wigglytuff', 'Charizard', 'Slowpoke', 'Poliwag', 'Tentacruel', 'Rhyhorn', 'Onix', 'Butterfree', 'Exeggcute', 'Sandslash', 'Pinsir', 'Rattata', 'Growlithe', 'Haunter', 'Pidgey', 'Ditto', 'Farfetchd', 'Pikachu', 'Raticate', 'Wartortle', 'Vaporeon', 'Cloyster', 'Hypno', 'Arbok', 'Metapod', 'Tangela', 'Kingler', 'Exeggutor', 'Kadabra', 'Seel', 'Voltorb', 'Chansey', 'Venomoth', 'Ponyta', 'Vileplume', 'Koffing', 'Blastoise', 'Tentacool', 'Lickitung', 'Paras', 'Clefable', 'Cubone', 'Marowak', 'Nidorino', 'Jolteon', 'Muk', 'Magikarp', 'Slowbro', 'Tauros', 'Kabuto', 'Spearow', 'Sandshrew', 'Eevee', 'Kakuna', 'Omastar', 'Ekans', 'Geodude', 'Magmar', 'Snorlax', 'Meowth', 'Pidgeotto', 'Venusaur', 'Persian', 'Rhydon', 'Starmie', 'Charmeleon', 'Lapras', 'Alakazam', 'Graveler', 'Psyduck', 'Rapidash', 'Doduo', 'Magneton', 'Arcanine', 'Electrode', 'Omanyte', 'Poliwhirl', 'Mew', 'Alolan Sandslash', 'Mewtwo', 'Weezing', 'Gastly', 'Victreebel', 'Ivysaur', 'MrMime', 'Shellder', 'Scyther', 'Diglett', 'Primeape', 'Raichu']
```
### Number of Images
```json
{'train': 4869, 'valid': 1390, 'test': 732}
```
### How to Use
- Install [datasets](https://pypi.org/project/datasets/):
```bash
pip install datasets
```
- Load the dataset:
```python
from datasets import load_dataset
ds = load_dataset("keremberke/pokemon-classification", name="full")
example = ds['train'][0]
```
### Roboflow Dataset Page
[https://universe.roboflow.com/robert-demo-qvail/pokedex/dataset/14](https://universe.roboflow.com/robert-demo-qvail/pokedex/dataset/14?ref=roboflow2huggingface)
### Citation
```
@misc{ pokedex_dataset,
title = { Pokedex Dataset },
type = { Open Source Dataset },
author = { Lance Zhang },
howpublished = { \\url{ https://universe.roboflow.com/robert-demo-qvail/pokedex } },
url = { https://universe.roboflow.com/robert-demo-qvail/pokedex },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2022 },
month = { dec },
note = { visited on 2023-01-14 },
}
```
### License
Public Domain
### Dataset Summary
This dataset was exported via roboflow.com on December 20, 2022 at 5:34 PM GMT
Roboflow is an end-to-end computer vision platform that helps you
* collaborate with your team on computer vision projects
* collect & organize images
* understand unstructured image data
* annotate, and create datasets
* export, train, and deploy computer vision models
* use active learning to improve your dataset over time
It includes 6991 images.
Pokemon are annotated in folder format.
The following pre-processing was applied to each image:
* Auto-orientation of pixel data (with EXIF-orientation stripping)
* Resize to 224x224 (Fit (black edges))
No image augmentation techniques were applied.
|
true |
# Dataset Card for The Schema-Guided Dialogue Dataset
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** [Github repository for The Schema-Guided Dialogue Dataset](https://github.com/google-research-datasets/dstc8-schema-guided-dialogue)
- **Paper:** [Towards Scalable Multi-Domain Conversational Agents: The Schema-Guided Dialogue Dataset](https://arxiv.org/abs/1909.05855)
- **Point of Contact:** [abhirast@google.com](abhirast@google.com)
### Dataset Summary
The Schema-Guided Dialogue dataset (SGD) was developed for the Dialogue State Tracking task of the Eights Dialogue Systems Technology Challenge (dstc8).
The SGD dataset consists of over 18k annotated multi-domain, task-oriented conversations between a human and a virtual assistant. These conversations involve interactions with services and APIs spanning 17 domains, ranging from banks and events to media, calendar, travel, and weather. For most of these domains, the SGD dataset contains multiple different APIs, many of which have overlapping functionalities but different interfaces, which reflects common real-world scenarios.
### Supported Tasks and Leaderboards
This dataset is designed to serve as an effective test-bed for intent prediction, slot filling, state tracking (i.e., estimating the user's goal) and language generation, among other tasks for large-scale virtual assistants:
- **Generative dialogue modeling** or `dialogue-modeling`: the text of the dialogues can be used to train a sequence model on the utterances. Performance on this task is typically evaluated with delexicalized-[BLEU](https://huggingface.co/metrics/bleu), inform rate and request success.
- **Intent state tracking**, a `multi-class-classification` task: predict the belief state of the user side of the conversation, performance is measured by [F1](https://huggingface.co/metrics/f1).
- **Action prediction**, a `parsing` task: parse an utterance into the corresponding dialog acts for the system to use. [F1](https://huggingface.co/metrics/f1) is typically reported.
### Languages
The text in the dataset is in English (`en`).
## Dataset Structure
### Data Instances
- `dialogues` configuration (default): Each dialogue is represented as a sequence of turns, each containing a user or system utterance. The annotations for each turn are grouped into frames, where each frame corresponds to a single service. The annotations for user turns include the active intent, the dialogue state and slot spans for the different slots values mentioned in the turn. For system turns, we have the system actions representing the semantics of the system utterance. Each system action is represented using a dialogue act with optional parameters.
- `schema` configuration: In addition to the dialogues, for each service used in the dataset, a normalized representation of the interface exposed is provided as the schema. The schema contains details like the name of the service, the list of tasks supported by the service (intents) and the attributes of the entities used by the service (slots). The schema also contains natural language descriptions of the service, intents and slots which can be used for developing models which can condition their predictions on the schema.
### Data Fields
Each dialog instance has the following fields:
- `dialogue_id`: A unique identifier for a dialogue.
- `services`: A list of services present in the dialogue.
- `turns`: A list of annotated system or user utterances. Each turn consists of the following fields:
- `speaker`: The speaker for the turn. Either `USER` or `SYSTEM`.
- `utterance`: A string containing the natural language utterance.
- `frames`: A list of frames, each frame containing annotations for a single service and consists of the following fields:
- `service`: The name of the service corresponding to the frame. The slots and intents used in the following fields are taken from the schema of this service.
- `slots`: A list of slot spans in the utterance, only provided for non-categorical slots. Each slot span contains the following fields:
- `slot`: The name of the slot.
- `start`: The index of the starting character in the utterance corresponding to the slot value.
- `exclusive_end`: The index of the character just after the last character corresponding to the slot value in the utterance.
- `actions`: A list of actions corresponding to the system. Each action has the following fields:
- `act`: The type of action.
- `slot`: (optional) A slot argument for some of the actions.
- `values`: (optional) A list of values assigned to the slot. If the values list is non-empty, then the slot must be present.
- `canonical_values`: (optional) The values in their canonicalized form as used by the service. It is a list of strings of the same length as values.
- `service_call`: (system turns only, optional) The request sent to the service. It consists of the following fields:
- `method`: The name of the intent or function of the service or API being executed.
- `parameters`: A pair of lists of the same lengths: `parameter_slot_name` contains slot names and `parameter_canonical_value` contains the corresponding values in their canonicalized form.
- `service_results`: (system turns only, optional) A list of entities containing the results obtained from the service. It is only available for turns in which a service call is made. Each entity is represented as a pair of lists of the same length: `service_slot_name` contains slot names and `service_canonical_value` contains the corresponding canonical values.
- `state`: (user turns only) The dialogue state corresponding to the service. It consists of the following fields:
- `active_intent`: The intent corresponding to the service of the frame which is currently being fulfilled by the system. It takes the value "NONE" if none of the intents are active.
- `requested_slots`: A list of slots requested by the user in the current turn.
- `slot_values`: A pair of lists of the same lengths: `slot_name` contains slot names and `slot_value_list` contains the corresponding lists of strings. For categorical slots, this list contains a single value assigned to the slot. For non-categorical slots, all the values in this list are spoken variations of each other and are equivalent (e.g, "6 pm", "six in the evening", "evening at 6" etc.).
The mapping from the action ID and the action name is the following:
0: AFFIRM
1: AFFIRM_INTENT
2: CONFIRM
3: GOODBYE
4: INFORM
5: INFORM_COUNT
6: INFORM_INTENT
7: NEGATE
8: NEGATE_INTENT
9: NOTIFY_FAILURE
10: NOTIFY_SUCCESS
11: OFFER
12: OFFER_INTENT
13: REQUEST
14: REQUEST_ALTS
15: REQ_MORE
16: SELECT
17: THANK_YOU
### Data Splits
The dataset is split into a `train`, `validation`, and `test` split with the following sizes:
| | train | validation | test |
|---------------------|------:|-----------:|------:|
| Number of dialogues | 16142 | 2482 | 4201 |
| Number of turns | 48426 | 7446 | 12603 |
## Dataset Creation
### Curation Rationale
The data was collected by first using a dialogue simulator to generate dialogue outlines first and then paraphrasing them to obtain natural utterances. Using a dialogue simulator ensures the coverage of a large variety of dialogue flows by filtering out similar flows in the simulation phase to create a diverse dataset, and dialogues can be generated with their annotation, as opposed to a Wizard-of-Oz setup which is prone to manual annotation errors.
### Source Data
#### Initial Data Collection and Normalization
The dialogue outlines are first generated by a simulator. The dialogue simulator interacts with the services to generate dialogue outlines. It consists of two
agents playing the roles of the user and the system, interacting with each other using a finite set of actions specified through dialogue acts over a probabilistic automaton designed to capture varied dialogue trajectories. It is worth noting that the simulation automaton does not include any domain-specific constraints: all domain-specific constraints are encoded in the schema and scenario.
The dialogue paraphrasing framework then converts the outlines generated by the simulator into a natural conversation. Users may refer to the slot values in the dialogue acts in various different ways during the conversation, e.g., “los angeles” may be referred to as “LA” or “LAX”. To introduce these natural variations in the slot values, different slot values are replaced with a randomly selected variation while being kept consistent across user turns in a dialogue. The actions are then converted to pseudo-natural language utterances using a set of manually defined action-to-text templates, and the resulting utterances for the different actions in a turn are concatenated together.
Finally, the dialogue transformed by these steps is sent to the crowd workers to be reformulated into more natural language. One crowd worker is tasked with paraphrasing all utterances of a dialogue to ensure naturalness and coherence. The crowd workers are asked to exactly repeat the slot values in their paraphrases so that the span indices for the slots can be recovered via string matching.
#### Who are the source language producers?
The language structure is machine-generated, and the language realizations are produced by crowd workers. The dataset paper does not provide demographic information for the crowd workers.
### Annotations
#### Annotation process
The annotations are automatically obtained during the initial sampling process and by string matching after reformulation.
#### Who are the annotators?
[N/A]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
The dataset was created by a team of researchers working at Google Mountain View.
### Licensing Information
The dataset is released under CC BY-SA 4.0 license.
### Citation Information
For the DSCT8 task, please cite:
```
@article{corr/abs-2002-01359,
author = {Abhinav Rastogi and
Xiaoxue Zang and
Srinivas Sunkara and
Raghav Gupta and
Pranav Khaitan},
title = {Schema-Guided Dialogue State Tracking Task at {DSTC8}},
journal = {CoRR},
volume = {abs/2002.01359},
year = {2020},
url = {https://arxiv.org/abs/2002.01359},
archivePrefix = {arXiv},
eprint = {2002.01359}
}
```
For the initial release paper please cite:
```
@inproceedings{aaai/RastogiZSGK20,
author = {Abhinav Rastogi and
Xiaoxue Zang and
Srinivas Sunkara and
Raghav Gupta and
Pranav Khaitan},
title = {Towards Scalable Multi-Domain Conversational Agents: The Schema-Guided
Dialogue Dataset},
booktitle = {The Thirty-Fourth {AAAI} Conference on Artificial Intelligence, {AAAI}
2020, The Thirty-Second Innovative Applications of Artificial Intelligence
Conference, {IAAI} 2020, The Tenth {AAAI} Symposium on Educational
Advances in Artificial Intelligence, {EAAI} 2020, New York, NY, USA,
February 7-12, 2020},
pages = {8689--8696},
publisher = {{AAAI} Press},
year = {2020},
url = {https://aaai.org/ojs/index.php/AAAI/article/view/6394}
}
```
### Contributions
Thanks to [@yjernite](https://github.com/yjernite) for adding this dataset. |
false |
# Dataset Card for fanpage
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-instances)
- [Data Splits](#data-instances)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** [Needs More Information]
- **Repository:** [Needs More Information]
- **Paper:** [Needs More Information]
- **Leaderboard:** [Needs More Information]
- **Point of Contact:** [Needs More Information]
### Dataset Summary
Fanpage dataset, containing news articles taken from Fanpage.
There are two features:
- source: Input news article.
- target: Summary of the article.
### Supported Tasks and Leaderboards
- `abstractive-summarization`, `summarization`
### Languages
The text in the dataset is in Italian
## Dataset Structure
### Data Instances
[Needs More Information]
### Data Fields
[Needs More Information]
### Data Splits
[Needs More Information]
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
More details and results in [published work](https://www.mdpi.com/2078-2489/13/5/228)
```
@Article{info13050228,
AUTHOR = {Landro, Nicola and Gallo, Ignazio and La Grassa, Riccardo and Federici, Edoardo},
TITLE = {Two New Datasets for Italian-Language Abstractive Text Summarization},
JOURNAL = {Information},
VOLUME = {13},
YEAR = {2022},
NUMBER = {5},
ARTICLE-NUMBER = {228},
URL = {https://www.mdpi.com/2078-2489/13/5/228},
ISSN = {2078-2489},
ABSTRACT = {Text summarization aims to produce a short summary containing relevant parts from a given text. Due to the lack of data for abstractive summarization on low-resource languages such as Italian, we propose two new original datasets collected from two Italian news websites with multi-sentence summaries and corresponding articles, and from a dataset obtained by machine translation of a Spanish summarization dataset. These two datasets are currently the only two available in Italian for this task. To evaluate the quality of these two datasets, we used them to train a T5-base model and an mBART model, obtaining good results with both. To better evaluate the results obtained, we also compared the same models trained on automatically translated datasets, and the resulting summaries in the same training language, with the automatically translated summaries, which demonstrated the superiority of the models obtained from the proposed datasets.},
DOI = {10.3390/info13050228}
}
``` |
false |
# MediQA
## Dataset Description
MEDIQA is a dataset of manually generated, question-driven summaries of multi and single document answers to consumer health questions.
- **Homepage:** https://osf.io/fyg46/?view_only=
### Citation Information
```
@article{savery2020question,
title={Question-driven summarization of answers to consumer health questions},
author={Savery, Max and Abacha, Asma Ben and Gayen, Soumya and Demner-Fushman, Dina},
journal={Scientific Data},
volume={7},
number={1},
pages={322},
year={2020},
publisher={Nature Publishing Group UK London}
}
``` |
false |
# Dataset Card for air_dialogue
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://worksheets.codalab.org/worksheets/0xa79833f4b3c24f4188cee7131b120a59
- **Repository:** https://github.com/google/airdialogue
- **Paper:** https://www.aclweb.org/anthology/D18-1419/
- **Leaderboard:** https://worksheets.codalab.org/worksheets/0xa79833f4b3c24f4188cee7131b120a59
- **Point of Contact:** [AirDialogue-Google](mailto:airdialogue@gmail.com)
[Aakash Gupta](mailto:aakashg80@gmail.com)
### Dataset Summary
AirDialogue, is a large dataset that contains 402,038 goal-oriented conversations. To collect this dataset, we create a contextgenerator which provides travel and flight restrictions. Then the human annotators are asked to play the role of a customer or an agent and interact with the goal of successfully booking a trip given the restrictions.
### Supported Tasks and Leaderboards
We use perplexity and BLEU score to evaluate the quality of the language generated by the model. We also compare the dialogue state generated by the model s and the ground truth state s0. Two categories of the metrics are used: exact match scores and scaled scores
The inference competition & leaderboard can be found here:
https://worksheets.codalab.org/worksheets/0xa79833f4b3c24f4188cee7131b120a59
### Languages
The text in the dataset is in English. The BCP 47 code is `en`
## Dataset Structure
### Data Instances
The data is provided in two set of files. The first one has the dialogues (`air_dialogue_data`) and the knowledge-base (`air_dialogue_kb`)
BuilderConfig: `air_dialogue_data`
```
{"action": {"status": "book", "name": "Emily Edwards", "flight": [1027]}, "intent": {"return_month": "June", "return_day": "14", "max_price": 200, "departure_airport": "DFW", "return_time": "afternoon", "max_connections": 1, "departure_day": "12", "goal": "book", "departure_month": "June", "name": "Emily Edwards", "return_airport": "IAD"}, "timestamps": [1519233239, 1519233244, 1519233249, 1519233252, 1519233333, 1519233374, 1519233392, 1519233416, 1519233443, 1519233448, 1519233464, 1519233513, 1519233525, 1519233540, 1519233626, 1519233628, 1519233638], "dialogue": ["customer: Hello.", "agent: Hello.", "customer: My name is Emily Edwards.", "agent: How may I help you out?", "customer: I need some help in my flight ticket reservation to attend a convocation meeting, can you please help me?", "agent: Sure, I will help you out. May I know your travelling dates please?", "customer: Thank you and my dates are 06/12 and back on 06/14.", "agent: Can I know your airport codes?", "customer: The airport codes are from DFW to IAD.", "agent: Ok, please wait a moment.", "customer: Sure.", "agent: There is a flight with connection 1 and price 200, can I proceed with this flight?", "customer: Yes, do proceed with booking.", "agent: Ok, your ticket has been booked.", "customer: Thank you for your assistance in my flight ticket reservation.", "agent: Thank you for choosing us.", "customer: You are welcome."], "expected_action": {"status": "book", "name": "Emily Edwards", "flight": [1027]}, "correct_sample": true}
```
BuilderConfig: `air_dialogue_kb`
```
{"kb": [{"return_airport": "DTW", "airline": "Spirit", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1000, "departure_month": "June", "departure_time_num": 17, "class": "economy", "return_time_num": 2, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 200}, {"return_airport": "DTW", "airline": "Frontier", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1001, "departure_month": "June", "departure_time_num": 0, "class": "business", "return_time_num": 15, "return_month": "June", "return_day": "13", "num_connections": 0, "price": 500}, {"return_airport": "DTW", "airline": "JetBlue", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1002, "departure_month": "June", "departure_time_num": 0, "class": "business", "return_time_num": 13, "return_month": "June", "return_day": "13", "num_connections": 1, "price": 600}, {"return_airport": "IAD", "airline": "Hawaiian", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1003, "departure_month": "June", "departure_time_num": 6, "class": "economy", "return_time_num": 5, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 200}, {"return_airport": "DFW", "airline": "AA", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1004, "departure_month": "June", "departure_time_num": 9, "class": "economy", "return_time_num": 11, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 100}, {"return_airport": "IAD", "airline": "AA", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1005, "departure_month": "June", "departure_time_num": 3, "class": "economy", "return_time_num": 17, "return_month": "June", "return_day": "13", "num_connections": 1, "price": 100}, {"return_airport": "DTW", "airline": "Frontier", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1006, "departure_month": "June", "departure_time_num": 10, "class": "economy", "return_time_num": 10, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 100}, {"return_airport": "IAD", "airline": "UA", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1007, "departure_month": "June", "departure_time_num": 14, "class": "economy", "return_time_num": 20, "return_month": "June", "return_day": "13", "num_connections": 1, "price": 100}, {"return_airport": "DFW", "airline": "AA", "departure_day": "13", "departure_airport": "DTW", "flight_number": 1008, "departure_month": "June", "departure_time_num": 6, "class": "economy", "return_time_num": 8, "return_month": "June", "return_day": "14", "num_connections": 2, "price": 400}, {"return_airport": "DFW", "airline": "Delta", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1009, "departure_month": "June", "departure_time_num": 18, "class": "economy", "return_time_num": 6, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 200}, {"return_airport": "DFW", "airline": "Frontier", "departure_day": "13", "departure_airport": "DTW", "flight_number": 1010, "departure_month": "June", "departure_time_num": 4, "class": "economy", "return_time_num": 2, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 100}, {"return_airport": "DFW", "airline": "Southwest", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1011, "departure_month": "June", "departure_time_num": 17, "class": "economy", "return_time_num": 22, "return_month": "June", "return_day": "13", "num_connections": 0, "price": 100}, {"return_airport": "DTW", "airline": "JetBlue", "departure_day": "11", "departure_airport": "DFW", "flight_number": 1012, "departure_month": "June", "departure_time_num": 13, "class": "economy", "return_time_num": 22, "return_month": "June", "return_day": "13", "num_connections": 1, "price": 100}, {"return_airport": "DTW", "airline": "Southwest", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1013, "departure_month": "June", "departure_time_num": 16, "class": "economy", "return_time_num": 13, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 200}, {"return_airport": "DTW", "airline": "Delta", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1014, "departure_month": "June", "departure_time_num": 0, "class": "economy", "return_time_num": 8, "return_month": "June", "return_day": "15", "num_connections": 1, "price": 100}, {"return_airport": "DTW", "airline": "Southwest", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1015, "departure_month": "June", "departure_time_num": 17, "class": "economy", "return_time_num": 1, "return_month": "June", "return_day": "15", "num_connections": 1, "price": 300}, {"return_airport": "DTW", "airline": "UA", "departure_day": "11", "departure_airport": "DFW", "flight_number": 1016, "departure_month": "June", "departure_time_num": 10, "class": "economy", "return_time_num": 4, "return_month": "June", "return_day": "14", "num_connections": 0, "price": 200}, {"return_airport": "DFW", "airline": "AA", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1017, "departure_month": "June", "departure_time_num": 14, "class": "economy", "return_time_num": 23, "return_month": "June", "return_day": "14", "num_connections": 2, "price": 400}, {"return_airport": "DTW", "airline": "JetBlue", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1018, "departure_month": "June", "departure_time_num": 3, "class": "economy", "return_time_num": 1, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 100}, {"return_airport": "DFW", "airline": "Hawaiian", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1019, "departure_month": "June", "departure_time_num": 7, "class": "economy", "return_time_num": 18, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 200}, {"return_airport": "DFW", "airline": "Delta", "departure_day": "12", "departure_airport": "IAD", "flight_number": 1020, "departure_month": "June", "departure_time_num": 6, "class": "economy", "return_time_num": 18, "return_month": "June", "return_day": "14", "num_connections": 2, "price": 200}, {"return_airport": "IAD", "airline": "Delta", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1021, "departure_month": "June", "departure_time_num": 11, "class": "business", "return_time_num": 8, "return_month": "June", "return_day": "14", "num_connections": 0, "price": 1000}, {"return_airport": "IAD", "airline": "JetBlue", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1022, "departure_month": "June", "departure_time_num": 4, "class": "economy", "return_time_num": 14, "return_month": "June", "return_day": "13", "num_connections": 0, "price": 200}, {"return_airport": "IAD", "airline": "Frontier", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1023, "departure_month": "June", "departure_time_num": 19, "class": "economy", "return_time_num": 23, "return_month": "June", "return_day": "13", "num_connections": 1, "price": 200}, {"return_airport": "DFW", "airline": "UA", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1024, "departure_month": "June", "departure_time_num": 11, "class": "economy", "return_time_num": 19, "return_month": "June", "return_day": "15", "num_connections": 1, "price": 200}, {"return_airport": "DTW", "airline": "Hawaiian", "departure_day": "11", "departure_airport": "IAD", "flight_number": 1025, "departure_month": "June", "departure_time_num": 6, "class": "economy", "return_time_num": 10, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 100}, {"return_airport": "DTW", "airline": "UA", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1026, "departure_month": "June", "departure_time_num": 0, "class": "economy", "return_time_num": 18, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 300}, {"return_airport": "IAD", "airline": "Delta", "departure_day": "12", "departure_airport": "DFW", "flight_number": 1027, "departure_month": "June", "departure_time_num": 17, "class": "economy", "return_time_num": 15, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 200}, {"return_airport": "IAD", "airline": "Southwest", "departure_day": "12", "departure_airport": "DTW", "flight_number": 1028, "departure_month": "June", "departure_time_num": 23, "class": "economy", "return_time_num": 13, "return_month": "June", "return_day": "14", "num_connections": 1, "price": 100}, {"return_airport": "DFW", "airline": "Spirit", "departure_day": "11", "departure_airport": "DTW", "flight_number": 1029, "departure_month": "June", "departure_time_num": 22, "class": "business", "return_time_num": 4, "return_month": "June", "return_day": "14", "num_connections": 0, "price": 800}], "reservation": 0}
```
### Data Fields
BuilderConfig: `air_dialogue_data`:
Provides for customer context, dialogue states and environment
key name | Description |
|---|---|
|'search_action' | search action performed by customer |
|'action' | Action taken by the agent |
|'intent' | Intents from the conversation |
|'timestamps' | Timestamp for each of the dialogues |
|'dialogue' | Dialogue recorded between agent & customer |
|'expected_action' | Expected action from agent (human-annotated)|
|'correct_sample' | whether action performed by agent was same as expected_action |
BuilderConfig: `air_dialogue_kb`:
Provides for the Agent Context _ca_ = (_db_, _r_ )
key name | Description |
|---|---|
|'kb' | Available flights in the database |
|'reservation' | whether customer has an existing reservation|
### Data Splits
Data is split into Train/Dev & Test in the ration of 80%, 10% and 10%
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
To collect this dataset, we create a contextgenerator which provides travel and flight restrictions. We then ask human annotators to play the role of a customer or an agent and interact with the goal of successfully booking a trip given the restrictions. Key to our environment is the ease of evaluating the success of the dialogue, which is achieved by using ground-truth states (e.g., the flight being booked) generated by the restrictions. Any dialogue agent that does not generate the correct states is considered to fail.
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
No personal and sensitive information is stored
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[AirDialogue team](mailto:airdialogue@gmail.com)
For issues regarding HuggingFace Dataset Hub implementation [Aakash Gupta](mailto:aakashg80@gmail.com)
### Licensing Information
cc-by-nc-4.0
### Citation Information
@inproceedings{wei-etal-2018-airdialogue,
title = "{A}ir{D}ialogue: An Environment for Goal-Oriented Dialogue Research",
author = "Wei, Wei and
Le, Quoc and
Dai, Andrew and
Li, Jia",
booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing",
month = oct # "-" # nov,
year = "2018",
address = "Brussels, Belgium",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/D18-1419",
doi = "10.18653/v1/D18-1419",
pages = "3844--3854",
abstract = "Recent progress in dialogue generation has inspired a number of studies on dialogue systems that are capable of accomplishing tasks through natural language interactions. A promising direction among these studies is the use of reinforcement learning techniques, such as self-play, for training dialogue agents. However, current datasets are limited in size, and the environment for training agents and evaluating progress is relatively unsophisticated. We present AirDialogue, a large dataset that contains 301,427 goal-oriented conversations. To collect this dataset, we create a context-generator which provides travel and flight restrictions. We then ask human annotators to play the role of a customer or an agent and interact with the goal of successfully booking a trip given the restrictions. Key to our environment is the ease of evaluating the success of the dialogue, which is achieved by using ground-truth states (e.g., the flight being booked) generated by the restrictions. Any dialogue agent that does not generate the correct states is considered to fail. Our experimental results indicate that state-of-the-art dialogue models can only achieve a score of 0.17 while humans can reach a score of 0.91, which suggests significant opportunities for future improvement.",
}
### Contributions
Thanks to [@skyprince999](https://github.com/skyprince999) for adding this dataset. |
true | # Dataset Card for Multilingual HateCheck
## Dataset Description
Multilingual HateCheck (MHC) is a suite of functional tests for hate speech detection models in 10 different languages: Arabic, Dutch, French, German, Hindi, Italian, Mandarin, Polish, Portuguese and Spanish.
For each language, there are 25+ functional tests that correspond to distinct types of hate and challenging non-hate.
This allows for targeted diagnostic insights into model performance.
For more details, please refer to our paper about MHC, published at the 2022 Workshop on Online Abuse and Harms (WOAH) at NAACL 2022. If you are using MHC, please cite our work!
- **Paper:** Röttger et al. (2022) - Multilingual HateCheck: Functional Tests for Multilingual Hate Speech Detection Models. https://arxiv.org/abs/2206.09917
- **Repository:** https://github.com/rewire-online/multilingual-hatecheck
- **Point of Contact:** paul@rewire.online
## Dataset Structure
The csv format mostly matches the original HateCheck data, with some adjustments for specific languages.
**mhc_case_id**
The test case ID that is unique to each test case across languages (e.g., "mandarin-1305")
**functionality**
The shorthand for the functionality tested by the test case (e.g, "target_obj_nh"). The same functionalities are tested in all languages, except for Mandarin and Arabic, where non-Latin script required adapting the tests for spelling variations.
**test_case**
The test case text.
**label_gold**
The gold standard label ("hateful" or "non-hateful") of the test case. All test cases within a given functionality have the same gold standard label.
**target_ident**
Where applicable, the protected group that is targeted or referenced in the test case. All HateChecks cover seven target groups, but their composition varies across languages.
**ref_case_id**
For hateful cases, where applicable, the ID of the hateful case which was perturbed to generate this test case. For non-hateful cases, where applicable, the ID of the hateful case which is contrasted by this test case.
**ref_templ_id**
The equivalent to ref_case_id, but for template IDs.
**templ_id**
The ID of the template from which the test case was generated.
**case_templ**
The template from which the test case was generated (where applicable).
**gender_male** and **gender_female**
For gender-inflected languages (French, Spanish, Portuguese, Hindi, Arabic, Italian, Polish, German), only for cases where gender inflection is relevant, separate entries for gender_male and gender_female replace case_templ.
**label_annotated**
A list of labels given by the three annotators who reviewed the test case (e.g., "['hateful', 'hateful', 'hateful']").
**label_annotated_maj**
The majority vote of the three annotators (e.g., "hateful"). In some cases this differs from the gold label given by our language experts.
**disagreement_in_case**
True if label_annotated_maj does not match label_gold for the entry.
**disagreement_in_template**
True if the test case is generated from an IDENT template and there is at least one case with disagreement_in_case generated from the same template. This can be used to exclude entire templates from MHC. |
true |
# Dataset Card for [financial-reports-sec]
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Dataset Configurations](#dataset-configurations)
- [Usage](#usage)
- [Supported Tasks](#supported-tasks)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Summary Statistics](#dataset-summary-statistics)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [References](#references)
- [Citation Information](#citation-information)
## Dataset Description
- **Point of Contact: Aman Khan**
### Dataset Summary
The dataset contains the annual report of US public firms filing with the SEC EDGAR system from 1993-2020. Each annual report (**10K filing**) is broken into 20 sections. Each section is split into individual sentences. Sentiment labels are provided on a **per filing basis** from the market reaction around the filing date for 3 different time windows _[t-1, t+1]_, _[t-1, t+5]_ and _[t-1, t+30]_. Additional metadata for each filing is included in the dataset.
### Dataset Configurations
**Four** configurations are available:
- _**large_lite**_:
- Contains only the basic features needed. Extra metadata is ommitted.
- Features List:
- **cik**
- **sentence**
- **section**
- **labels**
- **filingDate**
- **docID**
- **sentenceID**
- **sentenceCount**
- _**large_full**_:
- All features are included.
- Features List (excluding those already in the lite verison above):
- **name**
- **tickers**
- **exchanges**
- **entityType**
- **sic**
- **stateOfIncorporation**
- **tickerCount**
- **acceptanceDateTime**
- **form**
- **reportDate**
- **returns**
- _**small_lite**_:
- Same as _**large_lite**_ version except that only (200,000/20,000/20,000) sentences are loaded for (train/test/validation) splits.
- _**small_full**_:
- Same as _**large_full**_ version except that only (200,000/20,000/20,000) sentences are loaded for (train/test/validation) splits.
### Usage
```python
import datasets
# Load the lite configuration of the dataset
raw_dataset = datasets.load_dataset("JanosAudran/financial-reports-sec", "large_lite")
# Load a specific split
raw_dataset = datasets.load_dataset("JanosAudran/financial-reports-sec", "small_full", split="train")
```
### Supported Tasks
The tasks the dataset can be used directly for includes:
- _Masked Language Modelling_
- A model like BERT can be fine-tuned on this corpus of financial text.
- _Sentiment Analysis_
- For each annual report a label ["positive", "negative"] is provided based on the market reaction around the filing date (refer to [Annotations](#annotations)).
- _Next Sentence Prediction/Sentence Order Prediction_
- Sentences extracted from the filings are in their original order and as such the dataset can be adapted very easily for either of these tasks.
### Languages
All sentences are in English.
## Dataset Structure
### Data Instances
Refer to dataset preview.
### Data Fields
**Feature Name**
- Description
- Data type
- Example/Structure
**cik**
- 10 digit identifier used by SEC for a firm.
- _string_
- '0000001750'
**sentence**
- A single sentence from the 10-K filing.
- _string_
- 'The finance agreement is secured by a first priority security interest in all insurance policies, all unearned premium, return premiums, dividend payments and loss payments thereof.'
**section**
- The section of the 10-K filing the sentence is located.
- _ClassLabel_
- ```python
ClassLabel(names=['section_1', 'section_10', 'section_11', 'section_12', 'section_13', 'section_14', 'section_15', 'section_1A', 'section_1B', 'section_2','section_3', 'section_4', 'section_5', 'section_6', 'section_7', 'section_7A','section_8', 'section_9', 'section_9A', 'section_9B'], id=None)
```
**labels**
- The sentiment label for the entire filing (_**positve**_ or _**negative**_) based on different time windows.
- _Dict of ClassLables_
- ```python
{
'1d': ClassLabel(names=['positive', 'negative'], id=None),
'5d': ClassLabel(names=['positive', 'negative'], id=None),
'30d': ClassLabel(names=['positive', 'negative'], id=None)
}
```
**filingDate**
- The date the 10-K report was filed with the SEC.
- _string_
- '2021-03-10'
**docID**
- Unique ID for identifying the exact 10-K filing. Unique across all configs and splits. Can be used to identify the document from which the sentence came from.
- _string_
- '0000001750_10-K_2020'
**sentenceID**
- Unique ID for identifying the exact sentence. Unique across all configs and splits.
- _string_
- '0000001750_10-K_2020_section_1_100'
**sentenceCount**
- Integer identiying the running sequence for the sentence. Unique **only** for a given config and split.
- _string_
- 123
**name**
- The name of the filing entity
- _string_
- 'Investar Holding Corp'
**tickers**
- List of ticker symbols for the filing entity.
- _List of strings_
- ['ISTR']
**exchanges**
- List of exchanges for the filing entity.
- _List of strings_
- ['Nasdaq']
**entityType**
- The type of entity as identified in the 10-K filing.
- _string_
- 'operating'
**sic**
- Four digit SIC code for the filing entity.
- _string_
- '6022'
**stateOfIncorporation**
- Two character code for the state of incorporation for the filing entity.
- _string_
- 'LA'
**tickerCount**
- _**Internal use**_. Count of ticker symbols. Always 1.
- _int_
- 1
**acceptanceDateTime**
- The full timestamp of when the filing was accepted into the SEC EDGAR system.
- _string_
- '2021-03-10T14:26:11.000Z'
**form**
- The type of filing. Always 10-K in the dataset.
- _string_
- '10-K'
**reportDate**
- The last date in the fiscal year for which the entity is filing the report.
- _string_
- '2020-12-31'
**returns**
- _**Internal use**_. The prices and timestamps used to calculate the sentiment labels.
- _Dict_
- ```python
{'1d': {
'closePriceEndDate': 21.45746421813965,
'closePriceStartDate': 20.64960479736328,
'endDate': '2021-03-11T00:00:00-05:00',
'startDate': '2021-03-09T00:00:00-05:00',
'ret': 0.03912226855754852
},
'5d': {
'closePriceEndDate': 21.743167877197266,
'closePriceStartDate': 20.64960479736328,
'endDate': '2021-03-15T00:00:00-04:00',
'startDate': '2021-03-09T00:00:00-05:00',
'ret': 0.052958063781261444
},
'30d': {
'closePriceEndDate': 20.63919448852539,
'closePriceStartDate': 20.64960479736328,
'endDate': '2021-04-09T00:00:00-04:00',
'startDate': '2021-03-09T00:00:00-05:00',
'ret': -0.0005041408003307879}}
```
### Data Splits
| Config | train | validation | test |
| ---------- | ---------: | ---------: | --------: |
| large_full | 67,316,227 | 1,585,561 | 2,965,174 |
| large_lite | 67,316,227 | 1,585,561 | 2,965,174 |
| small_full | 200,000 | 20,000 | 20,000 |
| small_lite | 200,000 | 20,000 | 20,000 |
### Dataset Summary Statistics
| Variable | count | mean | std | min | 1% | 25% | 50% | 75% | 99% | max |
| :-------------------------------- | ---------: | ----: | -----: | -----: | -----: | -----: | ----: | ----: | ----: | --------: |
| Unique Firm Count | 4,677 | | | | | | | | | |
| Filings Count | 55,349 | | | | | | | | | |
| Sentence Count | 71,866,962 | | | | | | | | | |
| Filings per Firm | 4,677 | 12 | 9 | 1 | 1 | 4 | 11 | 19 | 27 | 28 |
| Return per Filing - 1d | 55,349 | 0.008 | 0.394 | -0.973 | -0.253 | -0.023 | 0 | 0.02 | 0.367 | 77.977 |
| Return per Filing - 5d | 55,349 | 0.013 | 0.584 | -0.99 | -0.333 | -0.034 | 0 | 0.031 | 0.5 | 100 |
| Return per Filing - 30d | 55,349 | 0.191 | 22.924 | -0.999 | -0.548 | -0.068 | 0.001 | 0.074 | 1 | 5,002.748 |
| Sentences per Filing | 55,349 | 1,299 | 654 | 0 | 110 | 839 | 1,268 | 1,681 | 3,135 | 8,286 |
| Sentences by Section - section_1 | 55,349 | 221 | 183 | 0 | 0 | 97 | 180 | 293 | 852 | 2,724 |
| Sentences by Section - section_10 | 55,349 | 24 | 40 | 0 | 0 | 4 | 6 | 20 | 173 | 1,594 |
| Sentences by Section - section_11 | 55,349 | 16 | 47 | 0 | 0 | 3 | 3 | 4 | 243 | 808 |
| Sentences by Section - section_12 | 55,349 | 9 | 14 | 0 | 0 | 3 | 4 | 8 | 56 | 1,287 |
| Sentences by Section - section_13 | 55,349 | 8 | 20 | 0 | 0 | 3 | 3 | 4 | 79 | 837 |
| Sentences by Section - section_14 | 55,349 | 22 | 93 | 0 | 0 | 3 | 3 | 8 | 413 | 3,536 |
| Sentences by Section - section_15 | 55,349 | 177 | 267 | 0 | 0 | 9 | 26 | 315 | 1104 | 4,140 |
| Sentences by Section - section_1A | 55,349 | 197 | 204 | 0 | 0 | 3 | 158 | 292 | 885 | 2,106 |
| Sentences by Section - section_1B | 55,349 | 4 | 31 | 0 | 0 | 1 | 3 | 3 | 13 | 2,414 |
| Sentences by Section - section_2 | 55,349 | 16 | 45 | 0 | 0 | 6 | 8 | 13 | 169 | 1,903 |
| Sentences by Section - section_3 | 55,349 | 14 | 36 | 0 | 0 | 4 | 5 | 12 | 121 | 2,326 |
| Sentences by Section - section_4 | 55,349 | 7 | 17 | 0 | 0 | 3 | 3 | 4 | 66 | 991 |
| Sentences by Section - section_5 | 55,349 | 20 | 41 | 0 | 0 | 10 | 15 | 21 | 87 | 3,816 |
| Sentences by Section - section_6 | 55,349 | 8 | 29 | 0 | 0 | 3 | 4 | 7 | 43 | 2,156 |
| Sentences by Section - section_7 | 55,349 | 265 | 198 | 0 | 0 | 121 | 246 | 373 | 856 | 4,539 |
| Sentences by Section - section_7A | 55,349 | 18 | 52 | 0 | 0 | 3 | 9 | 21 | 102 | 3,596 |
| Sentences by Section - section_8 | 55,349 | 257 | 296 | 0 | 0 | 3 | 182 | 454 | 1105 | 4,431 |
| Sentences by Section - section_9 | 55,349 | 5 | 33 | 0 | 0 | 3 | 3 | 4 | 18 | 2,330 |
| Sentences by Section - section_9A | 55,349 | 17 | 16 | 0 | 0 | 8 | 15 | 23 | 50 | 794 |
| Sentences by Section - section_9B | 55,349 | 4 | 18 | 0 | 0 | 2 | 3 | 4 | 23 | 813 |
| Word count per Sentence | 71,866,962 | 28 | 22 | 1 | 2 | 16 | 24 | 34 | 98 | 8,675 |
## Dataset Creation
### Curation Rationale
To create this dataset multiple sources of information have to be cleaned and processed for data merging. Starting from the raw filings:
- Useful metadata about the filing and firm was added.
- Time windows around the filing date were carefully created.
- Stock price data was then added for the windows.
- Ambiguous/duplicate records were removed.
### Source Data
#### Initial Data Collection and Normalization
Initial data was collected and processed by the authors of the research paper [**EDGAR-CORPUS: Billions of Tokens Make The World Go Round**](#references). Market price and returns data was collected from Yahoo Finance. Additional metadata was collected from SEC.
#### Who are the source language producers?
US public firms filing with the SEC.
### Annotations
#### Annotation process
Labels for sentiment classification are based on buy-and-hold returns over a fixed time window around the filing date with the SEC i.e. when the data becomes public. Returns are chosen for this process as it reflects the combined market intelligence at parsing the new information in the filings. For each filing date **t** the stock price at **t-1** and **t+W** is used to calculate returns. If, the returns are positive a label of **positive** is assigned else a label of **negative** is assigned. Three different windows are used to assign the labels:
- **1d**: _[t-1, t+1]_
- **5d**: _[t-1, t+5]_
- **30d**: _[t-1, t+30]_
The windows are based on calendar days and are adjusted for weekends and holidays. The rationale behind using 3 windows is as follows:
- A very short window may not give enough time for all the information contained in the filing to be reflected in the stock price.
- A very long window may capture other events that drive stock price for the firm.
#### Who are the annotators?
Financial market participants.
### Personal and Sensitive Information
The dataset contains public filings data from SEC. Market returns data was collected from Yahoo Finance.
## Considerations for Using the Data
### Social Impact of Dataset
Low to none.
### Discussion of Biases
The dataset is about financial information of public companies and as such the tone and style of text is in line with financial literature.
### Other Known Limitations
NA
## Additional Information
### Dataset Curators
**Aman Khan**
### Licensing Information
This dataset is provided under Apache 2.0.
### References
- Lefteris Loukas, Manos Fergadiotis, Ion Androutsopoulos, & Prodromos Malakasiotis. (2021). EDGAR-CORPUS [Data set]. Zenodo. https://doi.org/10.5281/zenodo.5589195
### Citation Information
Please use the following to cite this dataset:
```
@ONLINE{financial-reports-sec,
author = "Aman Khan",
title = "Financial Reports SEC",
url = "https://huggingface.co/datasets/JanosAudran/financial-reports-sec"
}
```
|
true |
# Dataset Card for "kor_nli"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://github.com/kakaobrain/KorNLUDatasets](https://github.com/kakaobrain/KorNLUDatasets)
- **Repository:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Paper:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Size of downloaded dataset files:** 126.34 MB
- **Size of the generated dataset:** 166.43 MB
- **Total amount of disk used:** 292.77 MB
### Dataset Summary
Korean Natural Language Inference datasets.
### Supported Tasks and Leaderboards
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Languages
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Dataset Structure
### Data Instances
#### multi_nli
- **Size of downloaded dataset files:** 42.11 MB
- **Size of the generated dataset:** 84.72 MB
- **Total amount of disk used:** 126.85 MB
An example of 'train' looks as follows.
```
```
#### snli
- **Size of downloaded dataset files:** 42.11 MB
- **Size of the generated dataset:** 80.13 MB
- **Total amount of disk used:** 122.25 MB
An example of 'train' looks as follows.
```
```
#### xnli
- **Size of downloaded dataset files:** 42.11 MB
- **Size of the generated dataset:** 1.56 MB
- **Total amount of disk used:** 43.68 MB
An example of 'validation' looks as follows.
```
```
### Data Fields
The data fields are the same among all splits.
#### multi_nli
- `premise`: a `string` feature.
- `hypothesis`: a `string` feature.
- `label`: a classification label, with possible values including `entailment` (0), `neutral` (1), `contradiction` (2).
#### snli
- `premise`: a `string` feature.
- `hypothesis`: a `string` feature.
- `label`: a classification label, with possible values including `entailment` (0), `neutral` (1), `contradiction` (2).
#### xnli
- `premise`: a `string` feature.
- `hypothesis`: a `string` feature.
- `label`: a classification label, with possible values including `entailment` (0), `neutral` (1), `contradiction` (2).
### Data Splits
#### multi_nli
| |train |
|---------|-----:|
|multi_nli|392702|
#### snli
| |train |
|----|-----:|
|snli|550152|
#### xnli
| |validation|test|
|----|---------:|---:|
|xnli| 2490|5010|
## Dataset Creation
### Curation Rationale
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
#### Who are the source language producers?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Annotations
#### Annotation process
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
#### Who are the annotators?
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Personal and Sensitive Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Discussion of Biases
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Other Known Limitations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Additional Information
### Dataset Curators
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Licensing Information
The dataset is licensed under Creative Commons [Attribution-ShareAlike license (CC BY-SA 4.0)](http://creativecommons.org/licenses/by-sa/4.0/).
### Citation Information
```
@article{ham2020kornli,
title={KorNLI and KorSTS: New Benchmark Datasets for Korean Natural Language Understanding},
author={Ham, Jiyeon and Choe, Yo Joong and Park, Kyubyong and Choi, Ilji and Soh, Hyungjoon},
journal={arXiv preprint arXiv:2004.03289},
year={2020}
}
```
### Contributions
Thanks to [@thomwolf](https://github.com/thomwolf), [@lhoestq](https://github.com/lhoestq), [@lewtun](https://github.com/lewtun), [@patrickvonplaten](https://github.com/patrickvonplaten) for adding this dataset. |
false |
## Dataset Description
- **Homepage:** https://www.wikiart.org/
### Dataset Summary
Dataset containing 81,444 pieces of visual art from various artists, taken from WikiArt.org,
along with class labels for each image :
* "artist" : 129 artist classes, including a "Unknown Artist" class
* "genre" : 11 genre classes, including a "Unknown Genre" class
* "style" : 27 style classes
On WikiArt.org, the description for the "Artworks by Genre" page reads :
A genre system divides artworks according to depicted themes and objects. A classical hierarchy of genres was developed in European culture by the 17th century. It ranked genres in high – history painting and portrait, - and low – genre painting, landscape and still life. This hierarchy was based on the notion of man as the measure of all things. Landscape and still life were the lowest because they did not involve human subject matter. History was highest because it dealt with the noblest events of humanity. Genre system is not so much relevant for a contemporary art; there are just two genre definitions that are usually applied to it: abstract or figurative.
The "Artworks by Style" page reads :
A style of an artwork refers to its distinctive visual elements, techniques and methods. It usually corresponds with an art movement or a school (group) that its author is associated with.
## Dataset Structure
* "image" : image
* "artist" : 129 artist classes, including a "Unknown Artist" class
* "genre" : 11 genre classes, including a "Unknown Genre" class
* "style" : 27 style classes
### Source Data
Files taken from this [archive](https://archive.org/download/wikiart-dataset/wikiart.tar.gz), curated from the [WikiArt website](https://www.wikiart.org/).
## Additional Information
Note:
* The WikiArt dataset can be used only for non-commercial research purpose.
* The images in the WikiArt dataset were obtained from WikiArt.org.
* The authors are neither responsible for the content nor the meaning of these images.
By using the WikiArt dataset, you agree to obey the terms and conditions of WikiArt.org.
### Contributions
[`gigant`](https://huggingface.co/gigant) added this dataset to the hub. |
false |
# Dataset Card for People's Speech
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-instances)
- [Data Splits](#data-instances)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** https://mlcommons.org/en/peoples-speech/
- **Repository:** https://github.com/mlcommons/peoples-speech
- **Paper:** https://arxiv.org/abs/2111.09344
- **Leaderboard:** [Needs More Information]
- **Point of Contact:** [datasets@mlcommons.org](mailto:datasets@mlcommons.org)
### Dataset Summary
The People's Speech Dataset is among the world's largest English speech recognition corpus today that is licensed for academic and commercial usage under CC-BY-SA and CC-BY 4.0. It includes 30,000+ hours of transcribed speech in English languages with a diverse set of speakers. This open dataset is large enough to train speech-to-text systems and crucially is available with a permissive license.
### Supported Tasks and Leaderboards
[Needs More Information]
### Languages
English
## Dataset Structure
### Data Instances
{
"id": "gov_DOT_uscourts_DOT_scotus_DOT_19-161/gov_DOT_uscourts_DOT_scotus_DOT_19-161_DOT_2020-03-02_DOT_mp3_00002.flac",
"audio": {
"path": "gov_DOT_uscourts_DOT_scotus_DOT_19-161/gov_DOT_uscourts_DOT_scotus_DOT_19-161_DOT_2020-03-02_DOT_mp3_00002.flac"
"array": array([-6.10351562e-05, ...]),
"sampling_rate": 16000
}
"duration_ms": 14490,
"text": "contends that the suspension clause requires a [...]"
}
### Data Fields
{
"id": datasets.Value("string"),
"audio": datasets.Audio(sampling_rate=16_000),
"duration_ms": datasets.Value("int32"),
"text": datasets.Value("string"),
}
### Data Splits
We provide the following configurations for the dataset: `cc-by-clean`, `cc-by-dirty`, `cc-by-sa-clean`, `cc-by-sa-dirty`, and `microset`. We don't provide splits for any of the configurations.
## Dataset Creation
### Curation Rationale
See our [paper](https://arxiv.org/abs/2111.09344).
### Source Data
#### Initial Data Collection and Normalization
Data was downloaded via the archive.org API. No data inference was done.
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
No manual annotation is done. We download only source audio with already existing transcripts.
#### Who are the annotators?
For the test and dev sets, we paid native American English speakers to do transcriptions. We do not know the identities of the transcriptionists for data in the training set. For the training set, we have noticed that some transcriptions are likely to be the output of automatic speech recognition systems.
### Personal and Sensitive Information
Several of our sources are legal and government proceedings, spoken histories, speeches, and so on. Given that these were intended as public documents and licensed as such, it is natural that the involved individuals are aware of this.
## Considerations for Using the Data
### Social Impact of Dataset
The dataset could be used for speech synthesis. However, this requires careful cleaning of the dataset, as background noise is not tolerable for speech synthesis.
The dataset could be used for keyword spotting tasks as well. In particular, this is good use case for the non-English audio in the dataset.
Our sincere hope is that the large breadth of sources our dataset incorporates reduces existing quality of service issues today, like speech recognition system’s poor understanding of non-native English accents. We cannot think of any unfair treatment that come from using this dataset at this time.
### Discussion of Biases
Our data is downloaded from archive.org. As such, the data is biased towards whatever users decide to upload there.
Almost all of our data is American accented English.
### Other Known Limitations
As of version 1.0, a portion of data in the training, test, and dev sets is poorly aligned. Specifically, some words appear in the transcript, but not the audio, or some words appear in the audio, but not the transcript. We are working on it.
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
We provide CC-BY and CC-BY-SA subsets of the dataset.
### Citation Information
Please cite:
```
@article{DBLP:journals/corr/abs-2111-09344,
author = {Daniel Galvez and
Greg Diamos and
Juan Ciro and
Juan Felipe Cer{\'{o}}n and
Keith Achorn and
Anjali Gopi and
David Kanter and
Maximilian Lam and
Mark Mazumder and
Vijay Janapa Reddi},
title = {The People's Speech: {A} Large-Scale Diverse English Speech Recognition
Dataset for Commercial Usage},
journal = {CoRR},
volume = {abs/2111.09344},
year = {2021},
url = {https://arxiv.org/abs/2111.09344},
eprinttype = {arXiv},
eprint = {2111.09344},
timestamp = {Mon, 22 Nov 2021 16:44:07 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2111-09344.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
``` |
false |
# Dataset Card for "QQP_triplets"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
- **Repository:** [More Information Needed](http://qim.fs.quoracdn.net/quora_duplicate_questions.tsv)
- **Paper:** [More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
- **Point of Contact:** [Kornél Csernai](https://www.quora.com/profile/Korn%C3%A9l-Csernai), [Nikhil Dandekar](https://www.quora.com/profile/Nikhil-Dandekar), [Shankar Iyer](https://www.quora.com/profile/Shankar-Iyer-5)
### Dataset Summary
This dataset will give anyone the opportunity to train and test models of semantic equivalence, based on actual Quora data. The data is organized as triplets (anchor, positive, negative).
Disclaimer: The team releasing Quora data did not upload the dataset to the Hub and did not write a dataset card.
These steps were done by the Hugging Face team.
### Supported Tasks
- [Sentence Transformers](https://huggingface.co/sentence-transformers) training; useful for semantic search and sentence similarity.
### Languages
- English.
## Dataset Structure
Each example is a dictionary with three keys (query, pos, and neg) containing a list each (triplets). The first key contains an anchor sentence, the second a positive sentence, and the third a list of negative sentences.
```
{"query": [anchor], "pos": [positive], "neg": [negative1, negative2, ..., negativeN]}
{"query": [anchor], "pos": [positive], "neg": [negative1, negative2, ..., negativeN]}
...
{"query": [anchor], "pos": [positive], "neg": [negative1, negative2, ..., negativeN]}
```
This dataset is useful for training Sentence Transformers models. Refer to the following post on how to train them.
### Usage Example
Install the 🤗 Datasets library with `pip install datasets` and load the dataset from the Hub with:
```python
from datasets import load_dataset
dataset = load_dataset("embedding-data/QQP_triplets")
```
The dataset is loaded as a `DatasetDict` and has the format:
```python
DatasetDict({
train: Dataset({
features: ['set'],
num_rows: 101762
})
})
```
Review an example `i` with:
```python
dataset["train"][i]["set"]
```
### Curation Rationale
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
#### Who are the source language producers?
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Annotations
#### Annotation process
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
#### Who are the annotators?
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Personal and Sensitive Information
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Discussion of Biases
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Other Known Limitations
Here are a few important things to keep in mind about this dataset:
- Our original sampling method returned an imbalanced dataset with many more true examples of duplicate pairs than non-duplicates.
Therefore, we supplemented the dataset with negative examples.
- One source of negative examples were pairs of “related questions” which, although pertaining to similar topics,
are not truly semantically equivalent.
- The distribution of questions in the dataset should not be taken to be representative of the distribution of questions asked on Quora. This is, in part, because of the combination of sampling procedures and also due to some sanitization measures that
have been applied to the final dataset (e.g., removal of questions with extremely long question details).
- The ground-truth labels contain some amount of noise: they are not guaranteed to be perfect.
## Additional Information
### Dataset Curators
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Licensing Information
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Citation Information
[More Information Needed](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)
### Contributions
Thanks to [Kornél Csernai](https://www.quora.com/profile/Korn%C3%A9l-Csernai), [Nikhil Dandekar](https://www.quora.com/profile/Nikhil-Dandekar), [Shankar Iyer](https://www.quora.com/profile/Shankar-Iyer-5) for adding this dataset.
|
false |
# Dataset Card for HAREM
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [HAREM homepage](https://www.linguateca.pt/primeiroHAREM/harem_coleccaodourada_en.html)
- **Repository:** [HAREM repository](https://www.linguateca.pt/primeiroHAREM/harem_coleccaodourada_en.html)
- **Paper:** [HAREM: An Advanced NER Evaluation Contest for Portuguese](http://comum.rcaap.pt/bitstream/10400.26/76/1/SantosSecoCardosoVilelaLREC2006.pdf)
- **Point of Contact:** [Diana Santos](mailto:diana.santos@sintef.no)
### Dataset Summary
The HAREM is a Portuguese language corpus commonly used for Named Entity Recognition tasks. It includes about 93k words, from 129 different texts,
from several genres, and language varieties. The split of this dataset version follows the division made by [1], where 7% HAREM
documents are the validation set and the miniHAREM corpus (with about 65k words) is the test set. There are two versions of the dataset set,
a version that has a total of 10 different named entity classes (Person, Organization, Location, Value, Date, Title, Thing, Event,
Abstraction, and Other) and a "selective" version with only 5 classes (Person, Organization, Location, Value, and Date).
It's important to note that the original version of the HAREM dataset has 2 levels of NER details, namely "Category" and "Sub-type".
The dataset version processed here ONLY USE the "Category" level of the original dataset.
[1] Souza, Fábio, Rodrigo Nogueira, and Roberto Lotufo. "BERTimbau: Pretrained BERT Models for Brazilian Portuguese." Brazilian Conference on Intelligent Systems. Springer, Cham, 2020.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
Portuguese
## Dataset Structure
### Data Instances
```
{
"id": "HAREM-871-07800",
"ner_tags": [3, 0, 0, 3, 4, 4, 4, 4, 4, 4, 4, 4,
],
"tokens": [
"Abraço", "Página", "Principal", "ASSOCIAÇÃO", "DE", "APOIO", "A", "PESSOAS", "COM", "VIH", "/", "SIDA"
]
}
```
### Data Fields
- `id`: id of the sample
- `tokens`: the tokens of the example text
- `ner_tags`: the NER tags of each token
The NER tags correspond to this list:
```
"O", "B-PESSOA", "I-PESSOA", "B-ORGANIZACAO", "I-ORGANIZACAO", "B-LOCAL", "I-LOCAL", "B-TEMPO", "I-TEMPO", "B-VALOR", "I-VALOR", "B-ABSTRACCAO", "I-ABSTRACCAO", "B-ACONTECIMENTO", "I-ACONTECIMENTO", "B-COISA", "I-COISA", "B-OBRA", "I-OBRA", "B-OUTRO", "I-OUTRO"
```
The NER tags have the same format as in the CoNLL shared task: a B denotes the first item of a phrase and an I any non-initial word.
### Data Splits
The data is split into train, validation and test set for each of the two versions (default and selective). The split sizes are as follow:
| Train | Val | Test |
| ------ | ----- | ---- |
| 121 | 8 | 128 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{santos2006harem,
title={Harem: An advanced ner evaluation contest for portuguese},
author={Santos, Diana and Seco, Nuno and Cardoso, Nuno and Vilela, Rui},
booktitle={quot; In Nicoletta Calzolari; Khalid Choukri; Aldo Gangemi; Bente Maegaard; Joseph Mariani; Jan Odjik; Daniel Tapias (ed) Proceedings of the 5 th International Conference on Language Resources and Evaluation (LREC'2006)(Genoa Italy 22-28 May 2006)},
year={2006}
}
```
### Contributions
Thanks to [@jonatasgrosman](https://github.com/jonatasgrosman) for adding this dataset. |
false |
# Dataset Card for doc2dial
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://doc2dial.github.io
- **Repository:** [Needs More Information]
- **Paper:** https://www.aclweb.org/anthology/2020.emnlp-main.652.pdf
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
Doc2dial is dataset of goal-oriented dialogues that are grounded in the associated documents. It includes over 4500 annotated conversations with an average of 14 turns that are grounded in over 450 documents from four domains. Compared to the prior document-grounded dialogue datasets this dataset covers a variety of dialogue scenes in information-seeking conversations.
### Supported Tasks and Leaderboards
> Supported Task: [Shared Task](https://doc2dial.github.io/workshop2021/shared.html) hosted by DialDoc21 at ACL.
> Leaderboard: [LINK](https://eval.ai/web/challenges/challenge-page/793)
### Languages
English
## Dataset Structure
### Data Instances
Sample data instance for `dialogue_domain` :
```
{
"dial_id": "9f44c1539efe6f7e79b02eb1b413aa43",
"doc_id": "Top 5 DMV Mistakes and How to Avoid Them#3_0",
"domain": "dmv",
"turns": [
{
"da": "query_condition",
"references": [
{
"sp_id": "4",
"label": "precondition"
}
],
"role": "user",
"turn_id": 1,
"utterance": "Hello, I forgot o update my address, can you help me with that?"
},
{
"da": "response_solution",
"references": [
{
"sp_id": "6",
"label": "solution"
},
{
"sp_id": "7",
"label": "solution"
},
{
"sp_id": "4",
"label": "references"
}
],
"role": "agent",
"turn_id": 2,
"utterance": "hi, you have to report any change of address to DMV within 10 days after moving. You should do this both for the address associated with your license and all the addresses associated with all your vehicles."
},
{
"da": "query_solution",
"references": [
{
"sp_id": "56",
"label": "solution"
},
{
"sp_id": "48",
"label": "references"
}
],
"role": "user",
"turn_id": 3,
"utterance": "Can I do my DMV transactions online?"
},
{
"da": "respond_solution",
"references": [
{
"sp_id": "56",
"label": "solution"
},
{
"sp_id": "48",
"label": "references"
}
],
"role": "agent",
"turn_id": 4,
"utterance": "Yes, you can sign up for MyDMV for all the online transactions needed."
},
{
"da": "query_condition",
"references": [
{
"sp_id": "48",
"label": "precondition"
}
],
"role": "user",
"turn_id": 5,
"utterance": "Thanks, and in case I forget to bring all of the documentation needed to the DMV office, what can I do?"
},
{
"da": "respond_solution",
"references": [
{
"sp_id": "49",
"label": "solution"
},
{
"sp_id": "50",
"label": "solution"
},
{
"sp_id": "52",
"label": "solution"
},
{
"sp_id": "48",
"label": "references"
}
],
"role": "agent",
"turn_id": 6,
"utterance": "This happens often with our customers so that's why our website and MyDMV are so useful for our customers. Just check if you can make your transaction online so you don't have to go to the DMV Office."
},
{
"da": "query_solution",
"references": [
{
"sp_id": "6",
"label": "solution"
},
{
"sp_id": "7",
"label": "solution"
},
{
"sp_id": "4",
"label": "references"
}
],
"role": "user",
"turn_id": 7,
"utterance": "Ok, and can you tell me again where should I report my new address?"
},
{
"da": "respond_solution",
"references": [
{
"sp_id": "6",
"label": "solution"
},
{
"sp_id": "7",
"label": "solution"
},
{
"sp_id": "4",
"label": "references"
}
],
"role": "agent",
"turn_id": 8,
"utterance": "Sure. Any change of address must be reported to the DMV, that's for the address associated with your license and any of your vehicles."
},
{
"da": "query_condition",
"references": [
{
"sp_id": "40",
"label": "precondition"
}
],
"role": "user",
"turn_id": 9,
"utterance": "Can you tell me more about Traffic points and their cost?"
},
{
"da": "respond_solution",
"references": [
{
"sp_id": "41",
"label": "solution"
},
{
"sp_id": "43",
"label": "solution"
},
{
"sp_id": "40",
"label": "references"
}
],
"role": "agent",
"turn_id": 10,
"utterance": "Traffic points is the system used by DMV to track dangerous drivers. The cost of the traffic points is independent of the DRA, so you get a separate charge based on the total points you accumulate."
}
]
}
```
Sample data instance for `document_domain` :
```
{
"doc_id": "Benefits Planner: Retirement | Online Calculator (WEP Version)#1_0",
"domain": "ssa",
"doc_html_raw": "<main class=\"content\" id=\"content\" role=\"main\">\n\n<section>\n\n<div>\n<h2>\nBenefits Planner: Retirement\n</h2>\n</div>\n</section>\n\n\n<section>\n\n<div>\n\n<div>\n\n\n</div>\n\n<article>\n<section>\n\n<h3>Online Calculator (WEP Version)</h3>\n<p>The calculator shown below allows you to estimate your Social Security benefit.\nHowever, for the most accurate estimates, <a>use the Detailed Calculator</a>.</p>\n<p>You need to enter all your past earnings\n, which are shown on your <a>online </a>.</p>\n\n<p>Please Note:</p>\n<ul class=\"browser-default\">\n<li>The Online Calculator is updated periodically<span>*</span> with new benefit increases and other benefit amounts. Therefore, it is likely that your benefit estimates in the future will differ from those calculated today.</li>\n<li>The Online Calculator works on PCs and Macs with Javascript enabled.</li>\n<li>Some browsers may not allow you to print the table below. </li>\n</ul>\n<p></p>\n\n<div>\nThe Online Calculator temporarily stores information on your local computer while your browser is open. To protect your personal information, you should close your browser after you have finished your estimate.\n</div>\n<p></p>\n\n<div>\n<p>Note: If your birthday is on January 1st, we figure your benefit as if your birthday was in the previous year.</p>\n<p>If you qualify for benefits as a Survivor, your <a>full retirement age for survivors benefits</a> may be different.</p></div>\n\n<div>\n</div></section></article></div></section></main>",
"doc_html_ts": "<main><section><div><h2 sent_id=\"1\" text_id=\"1\">Benefits Planner: Retirement</h2></div></section><section><div><article><section><h3 sent_id=\"2\" text_id=\"2\">Online Calculator (WEP Version)</h3><div tag_id=\"1\"><u sent_id=\"3\" tag_id=\"1\"><u sent_id=\"3\" tag_id=\"1\" text_id=\"3\">The calculator shown below allows you to estimate your Social Security benefit .</u></u><u sent_id=\"4\" tag_id=\"1\"><u sent_id=\"4\" tag_id=\"1\" text_id=\"4\">However ,</u><u sent_id=\"4\" tag_id=\"1\" text_id=\"5\">for the most accurate estimates ,</u><u sent_id=\"4\" tag_id=\"1\" text_id=\"6\">use the Detailed Calculator .</u></u></div><div tag_id=\"2\"><u sent_id=\"5\" tag_id=\"2\"><u sent_id=\"5\" tag_id=\"2\" text_id=\"7\">You need to enter all your past earnings , which are shown on your online .</u></u></div><div tag_id=\"3\"><u sent_id=\"6\" tag_id=\"3\"><u sent_id=\"6\" tag_id=\"3\" text_id=\"8\">Please Note:</u></u></div><ul class=\"browser-default\" tag_id=\"3\"><li tag_id=\"3\"><div tag_id=\"3\"><u sent_id=\"9\" tag_id=\"3\"><u sent_id=\"9\" tag_id=\"3\" text_id=\"9\">The Online Calculator is updated periodically * with new benefit increases and other benefit amounts .</u></u><u sent_id=\"10\" tag_id=\"3\"><u sent_id=\"10\" tag_id=\"3\" text_id=\"10\">Therefore ,</u><u sent_id=\"10\" tag_id=\"3\" text_id=\"11\">it is likely that your benefit estimates in the future will differ from those calculated today .</u></u></div></li><li tag_id=\"3\"><u sent_id=\"11\" tag_id=\"3\"><u sent_id=\"11\" tag_id=\"3\" text_id=\"12\">The Online Calculator works on PCs and Macs with Javascript enabled .</u></u></li><li tag_id=\"3\"><u sent_id=\"12\" tag_id=\"3\"><u sent_id=\"12\" tag_id=\"3\" text_id=\"13\">Some browsers may not allow you to print the table below .</u></u></li></ul><div>The Online Calculator temporarily stores information on your local computer while your browser is open. To protect your personal information, you should close your browser after you have finished your estimate.</div><div><div tag_id=\"4\"><u sent_id=\"13\" tag_id=\"4\"><u sent_id=\"13\" tag_id=\"4\" text_id=\"14\">Note:</u></u><u sent_id=\"14\" tag_id=\"4\"><u sent_id=\"14\" tag_id=\"4\" text_id=\"15\">If your birthday is on January 1st ,</u><u sent_id=\"14\" tag_id=\"4\" text_id=\"16\">we figure your benefit as if your birthday was in the previous year .</u></u></div><div tag_id=\"5\"><u sent_id=\"15\" tag_id=\"5\"><u sent_id=\"15\" tag_id=\"5\" text_id=\"17\">If you qualify for benefits as a Survivor ,</u><u sent_id=\"15\" tag_id=\"5\" text_id=\"18\">your full retirement age for survivors benefits may be different .</u></u></div></div></section></article></div></section></main>",
"doc_text": "\n\nBenefits Planner: Retirement \n\n\nOnline Calculator (WEP Version) \nThe calculator shown below allows you to estimate your Social Security benefit. However , for the most accurate estimates , use the Detailed Calculator. You need to enter all your past earnings, which are shown on your online. Please Note: The Online Calculator is updated periodically * with new benefit increases and other benefit amounts. Therefore , it is likely that your benefit estimates in the future will differ from those calculated today. The Online Calculator works on PCs and Macs with Javascript enabled. Some browsers may not allow you to print the table below. Note: If your birthday is on January 1st , we figure your benefit as if your birthday was in the previous year. If you qualify for benefits as a Survivor , your full retirement age for survivors benefits may be different. ",
"title": "Benefits Planner: Retirement | Online Calculator (WEP Version)#1",
"spans": [
{
"end_sec": 32,
"end_sp": 32,
"id_sec": "t_0",
"id_sp": "1",
"parent_titles": "[]",
"start_sec": 0,
"start_sp": 0,
"tag": "h2",
"text_sec": "\n\nBenefits Planner: Retirement \n",
"text_sp": "\n\nBenefits Planner: Retirement \n",
"title": "Benefits Planner: Retirement"
},
{
"end_sec": 67,
"end_sp": 67,
"id_sec": "t_1",
"id_sp": "2",
"parent_titles": "[{'id_sp': '1', 'text': 'Benefits Planner: Retirement', 'level': 'h2'}]",
"start_sec": 32,
"start_sp": 32,
"tag": "h3",
"text_sec": "\n\nOnline Calculator (WEP Version) \n",
"text_sp": "\n\nOnline Calculator (WEP Version) \n",
"title": "Online Calculator (WEP Version)"
},
{
"end_sec": 220,
"end_sp": 147,
"id_sec": "1",
"id_sp": "3",
"parent_titles": "[]",
"start_sec": 67,
"start_sp": 67,
"tag": "u",
"text_sec": "The calculator shown below allows you to estimate your Social Security benefit. However , for the most accurate estimates , use the Detailed Calculator. ",
"text_sp": "The calculator shown below allows you to estimate your Social Security benefit. ",
"title": "Online Calculator (WEP Version)"
}
]
}
```
Sample data instance for `doc2dial_rc` :
```
{
"id": "78f72b08b43791a4a70363fe62b8de08_1",
"is_impossible": false,
"question": "Hello, I want to know about the retirement plan.",
"answers": {
"answer_start": [
0
],
"text": [
"\n\nBenefits Planner: Retirement \n\n\nOnline Calculator (WEP Version) \n"
]
},
"context": "\n\nBenefits Planner: Retirement \n\n\nOnline Calculator (WEP Version) \nThe calculator shown below allows you to estimate your Social Security benefit. However , for the most accurate estimates , use the Detailed Calculator. You need to enter all your past earnings, which are shown on your online. Please Note: The Online Calculator is updated periodically * with new benefit increases and other benefit amounts. Therefore , it is likely that your benefit estimates in the future will differ from those calculated today. The Online Calculator works on PCs and Macs with Javascript enabled. Some browsers may not allow you to print the table below. Note: If your birthday is on January 1st , we figure your benefit as if your birthday was in the previous year. If you qualify for benefits as a Survivor , your full retirement age for survivors benefits may be different. ",
"title": "Benefits Planner: Retirement | Online Calculator (WEP Version)#1_0",
"domain": "ssa"
}
```
### Data Fields
For `document_domain`,
- `doc_id`: the ID of a document;
- `title`: the title of the document;
- `domain`: the domain of the document;
- `doc_text`: the text content of the document (without HTML markups);
- `doc_html_ts`: the document content with HTML markups and the annotated spans that are indicated by `text_id` attribute, which corresponds to `id_sp`.
- `doc_html_raw`: the document content with HTML markups and without span annotations.
- `spans`: key-value pairs of all spans in the document, with `id_sp` as key. Each span includes the following,
- `id_sp`: the id of a span as noted by `text_id` in `doc_html_ts`;
- `start_sp`/ `end_sp`: the start/end position of the text span in `doc_text`;
- `text_sp`: the text content of the span.
- `id_sec`: the id of the (sub)section (e.g. `<p>`) or title (`<h2>`) that contains the span.
- `start_sec` / `end_sec`: the start/end position of the (sub)section in `doc_text`.
- `text_sec`: the text of the (sub)section.
- `title`: the title of the (sub)section.
- `parent_titles`: the parent titles of the `title`.
For `dialogue_domain`:
- `dial_id`: the ID of a dialogue;
- `doc_id`: the ID of the associated document;
- `domain`: domain of the document;
- `turns`: a list of dialogue turns. Each turn includes,
- `turn_id`: the time order of the turn;
- `role`: either "agent" or "user";
- `da`: dialogue act;
- `references`: the grounding span (`id_sp`) in the associated document. If a turn is an irrelevant turn, i.e., `da` ends with "ood", `reference` is empty. **Note** that spans with labels "*precondition*"/"*solution*" are the actual grounding spans. Spans with label "*reference*" are the related titles or contextual reference, which is used for the purpose of describing a dialogue scene better to crowd contributors.
- `utterance`: the human-generated utterance based on the dialogue scene.
For `doc2dial_rc`, this conforms to [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) data format. For how to load Doc2Dial data for reading comprehension task, please refer [here](https://github.com/doc2dial/sharedtask-dialdoc2021).
- `id`: the ID of a QA instance;
- `question`: user query;
- `answers`: the answers that are grounded in the associated document;
- `answer_start`: the start position of the grounding span in the associated document (`context`);
- `text`: the text content of the grounding span;
- `title`: the title of the associated document;
- `domain`: the domain of the associated document;
- `context`: the text content of the associated document (without HTML markups).
### Data Splits
Training & dev split for dialogue domain
Training split only for document domain
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Song Feng, Hui Wan, Chulaka Gunasekara, Siva Sankalp Patel,Sachindra Joshi. Luis A. Lastras
### Licensing Information
Creative Commons Attribution 3.0 Unported
### Citation Information
@inproceedings{feng-etal-2020-doc2dial,
title = "doc2dial: A Goal-Oriented Document-Grounded Dialogue Dataset",
author = "Feng, Song and Wan, Hui and Gunasekara, Chulaka and Patel, Siva and Joshi, Sachindra and Lastras, Luis",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.652",
}
### Contributions
Thanks to [@songfeng](https://github.com/songfeng), [@KMFODA](https://github.com/KMFODA) for adding this dataset. |
false |
# Dataset Card for arXiv Dataset
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Kaggle arXiv Dataset Homepage](https://www.kaggle.com/Cornell-University/arxiv)
- **Repository:**
- **Paper:** [On the Use of ArXiv as a Dataset](https://arxiv.org/abs/1905.00075)
- **Leaderboard:**
- **Point of Contact:** [Matt Bierbaum](mailto:matt.bierbaum@gmail.com)
### Dataset Summary
A dataset of 1.7 million arXiv articles for applications like trend analysis, paper recommender engines, category prediction, co-citation networks, knowledge graph construction and semantic search interfaces.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
The language supported is English
## Dataset Structure
### Data Instances
This dataset is a mirror of the original ArXiv data. Because the full dataset is rather large (1.1TB and growing), this dataset provides only a metadata file in the json format. An example is given below
```
{'id': '0704.0002',
'submitter': 'Louis Theran',
'authors': 'Ileana Streinu and Louis Theran',
'title': 'Sparsity-certifying Graph Decompositions',
'comments': 'To appear in Graphs and Combinatorics',
'journal-ref': None,
'doi': None,
'report-no': None,
'categories': 'math.CO cs.CG',
'license': 'http://arxiv.org/licenses/nonexclusive-distrib/1.0/',
'abstract': ' We describe a new algorithm, the $(k,\\ell)$-pebble game with colors, and use\nit obtain a characterization of the family of $(k,\\ell)$-sparse graphs and\nalgorithmic solutions to a family of problems concerning tree decompositions of\ngraphs. Special instances of sparse graphs appear in rigidity theory and have\nreceived increased attention in recent years. In particular, our colored\npebbles generalize and strengthen the previous results of Lee and Streinu and\ngive a new proof of the Tutte-Nash-Williams characterization of arboricity. We\nalso present a new decomposition that certifies sparsity based on the\n$(k,\\ell)$-pebble game with colors. Our work also exposes connections between\npebble game algorithms and previous sparse graph algorithms by Gabow, Gabow and\nWestermann and Hendrickson.\n',
'update_date': '2008-12-13'}
```
### Data Fields
- `id`: ArXiv ID (can be used to access the paper)
- `submitter`: Who submitted the paper
- `authors`: Authors of the paper
- `title`: Title of the paper
- `comments`: Additional info, such as number of pages and figures
- `journal-ref`: Information about the journal the paper was published in
- `doi`: [Digital Object Identifier](https://www.doi.org)
- `report-no`: Report Number
- `abstract`: The abstract of the paper
- `categories`: Categories / tags in the ArXiv system
### Data Splits
The data was not splited.
## Dataset Creation
### Curation Rationale
For nearly 30 years, ArXiv has served the public and research communities by providing open access to scholarly articles, from the vast branches of physics to the many subdisciplines of computer science to everything in between, including math, statistics, electrical engineering, quantitative biology, and economics. This rich corpus of information offers significant, but sometimes overwhelming depth. In these times of unique global challenges, efficient extraction of insights from data is essential. To help make the arXiv more accessible, a free, open pipeline on Kaggle to the machine-readable arXiv dataset: a repository of 1.7 million articles, with relevant features such as article titles, authors, categories, abstracts, full text PDFs, and more is presented to empower new use cases that can lead to the exploration of richer machine learning techniques that combine multi-modal features towards applications like trend analysis, paper recommender engines, category prediction, co-citation networks, knowledge graph construction and semantic search interfaces.
### Source Data
This data is based on arXiv papers.
[More Information Needed]
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
This dataset contains no annotations.
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
The original data is maintained by [ArXiv](https://arxiv.org/)
### Licensing Information
The data is under the [Creative Commons CC0 1.0 Universal Public Domain Dedication](https://creativecommons.org/publicdomain/zero/1.0/)
### Citation Information
```
@misc{clement2019arxiv,
title={On the Use of ArXiv as a Dataset},
author={Colin B. Clement and Matthew Bierbaum and Kevin P. O'Keeffe and Alexander A. Alemi},
year={2019},
eprint={1905.00075},
archivePrefix={arXiv},
primaryClass={cs.IR}
}
```
### Contributions
Thanks to [@tanmoyio](https://github.com/tanmoyio) for adding this dataset. |
false |
# Wikipedia (simple English) embedded with cohere.ai `multilingual-22-12` encoder
We encoded [Wikipedia (simple English)](https://simple.wikipedia.org) using the [cohere.ai](https://txt.cohere.ai/multilingual/) `multilingual-22-12` embedding model.
To get an overview how this dataset was created and pre-processed, have a look at [Cohere/wikipedia-22-12](https://huggingface.co/datasets/Cohere/wikipedia-22-12).
## Embeddings
We compute for `title+" "+text` the embeddings using our `multilingual-22-12` embedding model, a state-of-the-art model that works for semantic search in 100 languages. If you want to learn more about this model, have a look at [cohere.ai multilingual embedding model](https://txt.cohere.ai/multilingual/).
## Further languages
We provide embeddings of Wikipedia in many different languages:
[ar](https://huggingface.co/datasets/Cohere/wikipedia-22-12-ar-embeddings), [de](https://huggingface.co/datasets/Cohere/wikipedia-22-12-de-embeddings), [en](https://huggingface.co/datasets/Cohere/wikipedia-22-12-en-embeddings), [es](https://huggingface.co/datasets/Cohere/wikipedia-22-12-es-embeddings), [fr](https://huggingface.co/datasets/Cohere/wikipedia-22-12-fr-embeddings), [hi](https://huggingface.co/datasets/Cohere/wikipedia-22-12-hi-embeddings), [it](https://huggingface.co/datasets/Cohere/wikipedia-22-12-it-embeddings), [ja](https://huggingface.co/datasets/Cohere/wikipedia-22-12-ja-embeddings), [ko](https://huggingface.co/datasets/Cohere/wikipedia-22-12-ko-embeddings), [simple english](https://huggingface.co/datasets/Cohere/wikipedia-22-12-simple-embeddings), [zh](https://huggingface.co/datasets/Cohere/wikipedia-22-12-zh-embeddings),
You can find the Wikipedia datasets without embeddings at [Cohere/wikipedia-22-12](https://huggingface.co/datasets/Cohere/wikipedia-22-12).
## Loading the dataset
You can either load the dataset like this:
```python
from datasets import load_dataset
docs = load_dataset(f"Cohere/wikipedia-22-12-simple-embeddings", split="train")
```
Or you can also stream it without downloading it before:
```python
from datasets import load_dataset
docs = load_dataset(f"Cohere/wikipedia-22-12-simple-embeddings", split="train", streaming=True)
for doc in docs:
docid = doc['id']
title = doc['title']
text = doc['text']
emb = doc['emb']
```
## Search
A full search example:
```python
#Run: pip install cohere datasets
from datasets import load_dataset
import torch
import cohere
co = cohere.Client(f"<<COHERE_API_KEY>>") # Add your cohere API key from www.cohere.com
#Load at max 1000 documents + embeddings
max_docs = 1000
docs_stream = load_dataset(f"Cohere/wikipedia-22-12-simple-embeddings", split="train", streaming=True)
docs = []
doc_embeddings = []
for doc in docs_stream:
docs.append(doc)
doc_embeddings.append(doc['emb'])
if len(docs) >= max_docs:
break
doc_embeddings = torch.tensor(doc_embeddings)
query = 'Who founded Youtube'
response = co.embed(texts=[query], model='multilingual-22-12')
query_embedding = response.embeddings
query_embedding = torch.tensor(query_embedding)
# Compute dot score between query embedding and document embeddings
dot_scores = torch.mm(query_embedding, doc_embeddings.transpose(0, 1))
top_k = torch.topk(dot_scores, k=3)
# Print results
print("Query:", query)
for doc_id in top_k.indices[0].tolist():
print(docs[doc_id]['title'])
print(docs[doc_id]['text'], "\n")
```
## Performance
You can find performance on the MIRACL dataset (a semantic search evaluation dataset) here: [miracl-en-queries-22-12#performance](https://huggingface.co/datasets/Cohere/miracl-en-queries-22-12#performance) |
false |
# Dataset Card for WikiDoc
For the dataset containing patient information from wikidoc refer to [this dataset](https://huggingface.co/datasets/medalpaca/medical_meadow_wikidoc_patient_information)
## Dataset Description
- **Source:** https://www.wikidoc.org/index.php/Main_Page
- **Repository:** https://github.com/kbressem/medalpaca
- **Paper:** TBA
### Dataset Summary
This dataset containes medical question-answer pairs extracted from [WikiDoc](https://www.wikidoc.org/index.php/Main_Page),
a collaborative platform for medical professionals to share and contribute to up-to-date medical knowledge.
The platform has to main subsites, the "Living Textbook" and "Patient Information". The "Living Textbook"
contains chapters for various medical specialties, which we crawled. We then used GTP-3.5-Turbo to rephrase
the paragraph heading to a question and used the paragraph as answer. Patient Information is structured differently,
in that each section subheading is already a question, making rephrasing them obsolete.
**Note:** This dataset is still a WIP. While the Q/A pairs from the patient information seems to be mostly correct,
the conversion using GPT-3.5-Turbo yielded some unsatisfactory results in approximately 30% of cases. We are in the process of cleaning this dataset.
### Citation Information
TBA |
false |
# Dataset Card for GEM/schema_guided_dialog
## Dataset Description
- **Homepage:** n/a
- **Repository:** [Github[(https://github.com/google-research-datasets/dstc8-schema-guided-dialogue)
- **Paper:** https://arxiv.org/abs/1909.05855
- **Leaderboard:** N/A
- **Point of Contact:** Abhinav Rastogi
### Link to Main Data Card
You can find the main data card on the [GEM Website](https://gem-benchmark.com/data_cards/schema_guided_dialog).
### Dataset Summary
The GEM version of this dataset functions as a response generation dataset. The input specifies dialog acts that a model needs to verbalize. The Schema-Guided Dialog dataset is challenging since it comprises multiple domains from hotel and travel to restaurants, and a wide range of dialog acts. The context of each conversation is provided as well.
You can load the dataset via:
```
import datasets
data = datasets.load_dataset('GEM/schema_guided_dialog')
```
The data loader can be found [here](https://huggingface.co/datasets/GEM/schema_guided_dialog).
#### website
n/a
#### paper
[Arxiv](https://arxiv.org/abs/1909.05855)
#### authors
Abhinav Rastogi, Xiaoxue Zang, Srinivas Sunkara, Raghav Gupta, Pranav Khaitan, Amir Fayazi, Maria Wang, and Guan-Lin Chao
## Dataset Overview
### Where to find the Data and its Documentation
#### Download
<!-- info: What is the link to where the original dataset is hosted? -->
<!-- scope: telescope -->
[Github[(https://github.com/google-research-datasets/dstc8-schema-guided-dialogue)
#### Paper
<!-- info: What is the link to the paper describing the dataset (open access preferred)? -->
<!-- scope: telescope -->
[Arxiv](https://arxiv.org/abs/1909.05855)
#### BibTex
<!-- info: Provide the BibTex-formatted reference for the dataset. Please use the correct published version (ACL anthology, etc.) instead of google scholar created Bibtex. -->
<!-- scope: microscope -->
```
{
@inproceedings{rastogi2020towards,
title={Towards scalable multi-domain conversational agents: The schema-guided dialogue dataset},
author={Rastogi, Abhinav and Zang, Xiaoxue and Sunkara, Srinivas and Gupta, Raghav and Khaitan, Pranav},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
volume={34},
number={05},
pages={8689--8696},
year={2020}
}
```
#### Contact Name
<!-- quick -->
<!-- info: If known, provide the name of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
Abhinav Rastogi
#### Contact Email
<!-- info: If known, provide the email of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
schema-guided-dst@google.com
#### Has a Leaderboard?
<!-- info: Does the dataset have an active leaderboard? -->
<!-- scope: telescope -->
no
### Languages and Intended Use
#### Multilingual?
<!-- quick -->
<!-- info: Is the dataset multilingual? -->
<!-- scope: telescope -->
no
#### Covered Languages
<!-- quick -->
<!-- info: What languages/dialects are covered in the dataset? -->
<!-- scope: telescope -->
`English`
#### Whose Language?
<!-- info: Whose language is in the dataset? -->
<!-- scope: periscope -->
The language structure is machine-generated, and the language realizations are produced by crowd workers.
The dataset paper does not provide demographic information for the crowd workers.
#### License
<!-- quick -->
<!-- info: What is the license of the dataset? -->
<!-- scope: telescope -->
cc-by-sa-4.0: Creative Commons Attribution Share Alike 4.0 International
#### Intended Use
<!-- info: What is the intended use of the dataset? -->
<!-- scope: microscope -->
The Schema-Guided Dialogue (SGD) dataset contains 18K multi-domain task-oriented dialogues between a human and a virtual assistant, which covers 17 domains ranging from banks and events to media, calendar, travel, and weather.
The language presents in the datset is only English.
The SGD dataset provides a challenging testbed for a number of tasks in task-oriented dialogue, including language understanding, slot filling, dialogue state tracking and response generation.
For the creation of the SGD dataset, they developed a multi-domain dialogue simulator that generates dialogue outlines over an arbitrary combination of APIs, dialogue states and system actions. Then, they used a crowd-sourcing procedure to paraphrase these outlines to natural language utterances.
This novel crowd-sourcing procedure preserves all annotations obtained from the simulator and does not require any extra annotations after dialogue collection.
#### Primary Task
<!-- info: What primary task does the dataset support? -->
<!-- scope: telescope -->
Dialog Response Generation
#### Communicative Goal
<!-- quick -->
<!-- info: Provide a short description of the communicative goal of a model trained for this task on this dataset. -->
<!-- scope: periscope -->
The goal of a speaker who generates the target utterance is to help users accomplish tasks including but not limited to finding flights, booking restaurants, searching for nearby events and movies.
### Credit
#### Curation Organization Type(s)
<!-- info: In what kind of organization did the dataset curation happen? -->
<!-- scope: telescope -->
`industry`
#### Curation Organization(s)
<!-- info: Name the organization(s). -->
<!-- scope: periscope -->
Google
#### Dataset Creators
<!-- info: Who created the original dataset? List the people involved in collecting the dataset and their affiliation(s). -->
<!-- scope: microscope -->
Abhinav Rastogi, Xiaoxue Zang, Srinivas Sunkara, Raghav Gupta, Pranav Khaitan, Amir Fayazi, Maria Wang, and Guan-Lin Chao
#### Funding
<!-- info: Who funded the data creation? -->
<!-- scope: microscope -->
Google
#### Who added the Dataset to GEM?
<!-- info: Who contributed to the data card and adding the dataset to GEM? List the people+affiliations involved in creating this data card and who helped integrate this dataset into GEM. -->
<!-- scope: microscope -->
Wanyu Du wrote the initial data card and Yacine Jernite the data loader. Simon Mille updated the data card with the additional splits. Sebastian Gehrmann migrated the data card and loader to the v2 version and extended the missing information.
### Dataset Structure
#### Data Fields
<!-- info: List and describe the fields present in the dataset. -->
<!-- scope: telescope -->
Each dialog instance has the following fields:
* `dialogue_id`: A unique identifier for a dialogue.
* `services`: A list of services present in the dialogue.
* `turns`: A list of annotated system or user utterances. Each turn consists of the following fields:
* `speaker`: The speaker for the turn, either `USER` or `SYSTEM`.
* `utterance`: A string containing the natural language utterance.
* `frames`: A list of frames, each frame containing annotations for a single service and consists of the following fields:
* `service`: The name of the service corresponding to the frame. The slots and intents used in the following fields are taken from the schema of this service.
* `slots`: A list of slot spans in the utterance, only provided for non-categorical slots. Each slot span contains the following fields:
* `slot`: The name of the slot.
* `start`: The index of the starting character in the utterance corresponding to the slot value.
* `exclusive_end`: The index of the character just after the last character corresponding to the slot value in the utterance.
* `actions`: A list of actions corresponding to the system. Each action has the following fields:
* `act`: The type of action.
* `slot`: (optional) A slot argument for some of the actions.
* `values`: (optional) A list of values assigned to the slot. If the values list is non-empty, then the slot must be present.
* `canonical_values`: (optional) The values in their canonicalized form as used by the service. It is a list of strings of the same length as values.
* `service_call`: (system turns only, optional) The request sent to the service. It consists of the following fields:
* `method`: The name of the intent or function of the service or API being executed.
* `parameters`: A pair of lists of the same lengths: `parameter_slot_name` contains slot names and `parameter_canonical_value` contains the corresponding values in their canonicalized form.
* `service_results`: (system turns only, optional) A list of entities containing the results obtained from the service. It is only available for turns in which a service call is made. Each entity is represented as a pair of lists of the same length: `service_slot_name` contains slot names and `service_canonical_value` contains the corresponding canonical values.
* `state`: (user turns only) The dialogue state corresponding to the service. It consists of the following fields:
* `active_intent`: The intent corresponding to the service of the frame which is currently being fulfilled by the system. It takes the value "NONE" if none of the intents are active.
* `requested_slots`: A list of slots requested by the user in the current turn.
* `slot_values`: A pair of lists of the same lengths: `slot_name` contains slot names and `slot_value_list` contains the corresponding lists of strings. For categorical slots, this list contains a single value assigned to the slot. For non-categorical slots, all the values in this list are spoken variations of each other and are equivalent (e.g, "6 pm", "six in the evening", "evening at 6" etc.).
#### Example Instance
<!-- info: Provide a JSON formatted example of a typical instance in the dataset. -->
<!-- scope: periscope -->
```
{'dialogue_id': '1_00000',
'services': ['Restaurants_1'],
'turns':
{'frames':
[{'actions': [{'act': [6],
'canonical_values': [['FindRestaurants']],
'slot': ['intent'],
'values': [['FindRestaurants']]}],
'service': ['Restaurants_1'],
'service_call': [{'method': '',
'parameters': {'parameter_canonical_value': [],
'parameter_slot_name': []}}],
'service_results': [{'service_results_list': []}],
'slots': [{'exclusive_end': [], 'slot': [], 'start': []}],
'state': [{'active_intent': 'FindRestaurants',
'requested_slots': [],
'slot_values': {'slot_name': [], 'slot_value_list': []}}]},
{'actions': [{'act': [13],
'canonical_values': [[]],
'slot': ['city'],
'values': [[]]}],
'service': ['Restaurants_1'],
'service_call': [{'method': '',
'parameters': {'parameter_canonical_value': [],
'parameter_slot_name': []}}],
'service_results': [{'service_results_list': []}],
'slots': [{'exclusive_end': [], 'slot': [], 'start': []}],
'state': [{'active_intent': '',
'requested_slots': [],
'slot_values': {'slot_name': [], 'slot_value_list': []}}]},
...,]}
'speaker': [0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1],
'utterance': [
'I am feeling hungry so I would like to find a place to eat.',
'Do you have a specific which you want the eating place to be located at?',
'I would like for it to be in San Jose.',
'Is there a specific cuisine type you enjoy, such as Mexican, Italian or something else?',
'I usually like eating the American type of food.',
'I see that at 71 Saint Peter there is a good restaurant which is in San Jose.',
'Can you give me the address of this restaurant.',
'If you want to go to this restaurant you can find it at 71 North San Pedro Street.',
'Can you give me the phone number that I can contact them with?',
'If you want to phone them you can at 408-971-8523.',
'Is there some other restaurant which you can suggest?',
'How would you like Bazille restaurant which is situated in San Jose.',
'Do you have another restaurant matching my needs? For example a restaurant which is economical and is located in Palo Alto.',
'I see that 7 restaurants suit to what you requested. Bird Dog seems as a good restaurant and is located in Palo Alto.',
'Alright, that seems good. I would like to make a booking at this restaurant.',
'For which time do you want the booking to be?',
'I will be eating there at 11:30 am so make it for then.',
'Can you please confirm that you want to book a table for 2 at 11:30 am at the Bird Dog restaurant in Palo Alto for today.',
'That suits me well. Can you tell me if they feature live music?',
'Your booking has been made without errors, but unfortunately they do not have live music.',
'Will I be able to find liquor there? Can you give me the address of their location?',
'The restaurant is located at 420 Ramona Street. Unfortunately they do not serve alcohol at the restaurant.',
'I appreciate it very much. That would be all.',
'Have a good time!'
]}
```
#### Data Splits
<!-- info: Describe and name the splits in the dataset if there are more than one. -->
<!-- scope: periscope -->
The dataset is split into a train, validation, and test set with the following sizes:
| | Train | Validation | Test |
| --- | --- | --- | --- |
| \# of dialogues | 16142 | 2482 | 4201 |
| \# of turns | 48426 | 7446 | 12603 |
#### Splitting Criteria
<!-- info: Describe any criteria for splitting the data, if used. If there are differences between the splits (e.g., if the training annotations are machine-generated and the dev and test ones are created by humans, or if different numbers of annotators contributed to each example), describe them here. -->
<!-- scope: microscope -->
The data is generally split i.i.d, but some topics only appear in training and some only for testing. For example, the domains Messaging, Payment, and Train are test-only.
## Dataset in GEM
### Rationale for Inclusion in GEM
#### Why is the Dataset in GEM?
<!-- info: What does this dataset contribute toward better generation evaluation and why is it part of GEM? -->
<!-- scope: microscope -->
This dataset comprises a wide range of dialog capabilities and thus enables the evaluation of many more generation capabilities of comparable datasets. Its collection methodology ensures a high diversity but also high quality of the data.
#### Similar Datasets
<!-- info: Do other datasets for the high level task exist? -->
<!-- scope: telescope -->
yes
#### Unique Language Coverage
<!-- info: Does this dataset cover other languages than other datasets for the same task? -->
<!-- scope: periscope -->
no
#### Difference from other GEM datasets
<!-- info: What else sets this dataset apart from other similar datasets in GEM? -->
<!-- scope: microscope -->
The domains a lot more diverse than other datasets.
#### Ability that the Dataset measures
<!-- info: What aspect of model ability can be measured with this dataset? -->
<!-- scope: periscope -->
surface realization, compositionality.
### GEM-Specific Curation
#### Modificatied for GEM?
<!-- info: Has the GEM version of the dataset been modified in any way (data, processing, splits) from the original curated data? -->
<!-- scope: telescope -->
yes
#### GEM Modifications
<!-- info: What changes have been made to he original dataset? -->
<!-- scope: periscope -->
`data points modified`
#### Modification Details
<!-- info: For each of these changes, described them in more details and provided the intended purpose of the modification -->
<!-- scope: microscope -->
We are focusing on the response-generation part of the dataset and thus reformatted the dataset to treat the service agent utterances as the targets to be generated and the previous customer utterance and the agent's dialog act as the input. We additionally reformat the dialog acts to directly conform to the format described in this [paper](https://arxiv.org/abs/2004.15006).
#### Additional Splits?
<!-- info: Does GEM provide additional splits to the dataset? -->
<!-- scope: telescope -->
yes
#### Split Information
<!-- info: Describe how the new splits were created -->
<!-- scope: periscope -->
9 challenge sets for Schema-Guided Dialog were added to the GEM evaluation suite.
1. We created subsets of the training and development sets of 500 randomly selected inputs each.
2. We applied 5 transformations to respectively 5 sets of 500 randomly selected inputs: (i) back-translation, (ii)-(iii) introduction of typographical errors, using Butterfingers with two thresholds (0.02 and 0.05), resulting in two sets with different amounts of typos introduced (there are more typos with the 0.05 threshold than with the 0.02 one), (iv) removal of final punctuations (when any), and (v) input scrambling, for which the order of the dialogue acts was randomly reassigned.
3. For the input size, we created subpopulations based on the number of dialogue acts in the input.
| DA number | Frequency English |
|---------------|-------------------|
| 1 | 5049 |
| 2 | 2517 |
| 3 | 1328 |
| 4 | 469 |
| 5 | 335 |
| 6 | 256 |
| 7 | 46 |
We also split the test data according to the type of dialogue act, represented by cardinal numbers in the dataset.
| DA type | Frequency English |
|--------------|-------------------|
| 2 | 1397 |
| 3 | 983 |
| 4 | 1027 |
| 5 | 958 |
| 9 | 72 |
| 10 | 1024 |
| 11 | 1246 |
| 12 | 500 |
| 13 | 2078 |
| 15 | 715 |
#### Split Motivation
<!-- info: What aspects of the model's generation capacities were the splits created to test? -->
<!-- scope: periscope -->
Generalization and Robustness.
### Getting Started with the Task
#### Pointers to Resources
<!-- info: Getting started with in-depth research on the task. Add relevant pointers to resources that researchers can consult when they want to get started digging deeper into the task. -->
<!-- scope: microscope -->
* [Paper for dataset and DST baseline](https://arxiv.org/pdf/1909.05855.pdf)
* [DSTC8 overview paper](https://arxiv.org/pdf/2002.01359.pdf)
* [Code for DST baseline](https://github.com/google-research/google-research/tree/master/schema_guided_dst)
* [Natural language generation baseline paper](https://arxiv.org/pdf/2004.15006.pdf)
* [Blog post announcing the dataset](https://ai.googleblog.com/2019/10/introducing-schema-guided-dialogue.html)
## Previous Results
### Previous Results
#### Measured Model Abilities
<!-- info: What aspect of model ability can be measured with this dataset? -->
<!-- scope: telescope -->
Surface realization and compositionally.
#### Metrics
<!-- info: What metrics are typically used for this task? -->
<!-- scope: periscope -->
`BLEURT`, `BLEU`, `ROUGE`
#### Proposed Evaluation
<!-- info: List and describe the purpose of the metrics and evaluation methodology (including human evaluation) that the dataset creators used when introducing this task. -->
<!-- scope: microscope -->
The original paper focused on the task of dialog state prediction instead of response generation and thus did not suggest any set of metrics.
#### Previous results available?
<!-- info: Are previous results available? -->
<!-- scope: telescope -->
no
## Dataset Curation
### Original Curation
#### Original Curation Rationale
<!-- info: Original curation rationale -->
<!-- scope: telescope -->
Previous multi-domain task-oriented dialogue datsets do not sufficiently capture the real-world challenges in virtual assistants, since they cover few domains and assume a single static ontology per domain.
The SGD datset is created to cover 17 domains with over 16K dialogues, and contain multiple different APIs in most domains, many of which have overlapping functionalities but different interfaces, which reflects common real-world scenarios.
The wide range of available annotations can be used for intent prediction, slot filling, dialogue state tracking, policy imitation learning, language generation, user simulation learning, among other tasks in large-scale virtual assistants.
#### Communicative Goal
<!-- info: What was the communicative goal? -->
<!-- scope: periscope -->
The goal of a speaker who generates the target utterance is to help users accomplish tasks including but not limited to finding flights, booking restaurants, searching for nearby events and movies.
#### Sourced from Different Sources
<!-- info: Is the dataset aggregated from different data sources? -->
<!-- scope: telescope -->
no
### Language Data
#### How was Language Data Obtained?
<!-- info: How was the language data obtained? -->
<!-- scope: telescope -->
`Machine-generated`
#### Generation Method Link
<!-- info: If text was machine-generated for the dataset, provide a link to the generation method if available (N/A otherwise). -->
<!-- scope: periscope -->
[Github](https://github.com/google-research-datasets/dstc8-schema-guided-dialogue)
#### Language Producers
<!-- info: What further information do we have on the language producers? -->
<!-- scope: microscope -->
The dialogue outlines are first generated by a simulator. The dialogue simulator interacts with the services to generate dialogue outlines. It consists of two agents playing the roles of the user and the system, interacting with each other using a finite set of actions specified through dialogue acts over a probabilistic automaton designed to capture varied dialogue trajectories. It is worth noting that the simulation automaton does not include any domain-specific constraints: all domain-specific constraints are encoded in the schema and scenario.
The dialogue paraphrasing framework then converts the outlines generated by the simulator into a natural conversation. Users may refer to the slot values in the dialogue acts in various different ways during the conversation, e.g., “los angeles” may be referred to as “LA” or “LAX”. To introduce these natural variations in the slot values, different slot values are replaced with a randomly selected variation while being kept consistent across user turns in a dialogue. The actions are then converted to pseudo-natural language utterances using a set of manually defined action-to-text templates, and the resulting utterances for the different actions in a turn are concatenated together.
#### Topics Covered
<!-- info: Does the language in the dataset focus on specific topics? How would you describe them? -->
<!-- scope: periscope -->
The dataset covers the following domains: Alarm, Banks, Buses, Calendar, Events, Flights, Homes, Hotels, Media, Messaging, Movies, Music, Payment, RentalCars, Restaurants, RideSharing, Services, Train, Travel, and Weather. The domain ‘Service’ includes salons, dentists, doctors etc. The ‘Alarm’, ‘Messaging’, ‘Payment’ and ‘Train’ domains are only present in the dev or test sets.
to test generalization to new domains.
#### Data Validation
<!-- info: Was the text validated by a different worker or a data curator? -->
<!-- scope: telescope -->
not validated
#### Was Data Filtered?
<!-- info: Were text instances selected or filtered? -->
<!-- scope: telescope -->
not filtered
### Structured Annotations
#### Additional Annotations?
<!-- quick -->
<!-- info: Does the dataset have additional annotations for each instance? -->
<!-- scope: telescope -->
crowd-sourced
#### Number of Raters
<!-- info: What is the number of raters -->
<!-- scope: telescope -->
unknown
#### Raters per Training Example
<!-- info: How many annotators saw each training example? -->
<!-- scope: periscope -->
0
#### Raters per Test Example
<!-- info: How many annotators saw each test example? -->
<!-- scope: periscope -->
0
#### Annotation Service?
<!-- info: Was an annotation service used? -->
<!-- scope: telescope -->
unknown
#### Annotation Values
<!-- info: Purpose and values for each annotation -->
<!-- scope: microscope -->
The dialogue transformed by these steps is sent to the crowd workers to be reformulated into more natural language. One crowd worker is tasked with paraphrasing all utterances of a dialogue to ensure naturalness and coherence. The crowd workers are asked to exactly repeat the slot values in their paraphrases so that the span indices for the slots can be recovered via string matching.
#### Any Quality Control?
<!-- info: Quality control measures? -->
<!-- scope: telescope -->
none
### Consent
#### Any Consent Policy?
<!-- info: Was there a consent policy involved when gathering the data? -->
<!-- scope: telescope -->
no
#### Justification for Using the Data
<!-- info: If not, what is the justification for reusing the data? -->
<!-- scope: microscope -->
While no policy is reported, we assume that one was in place for the collection.
### Private Identifying Information (PII)
#### Contains PII?
<!-- quick -->
<!-- info: Does the source language data likely contain Personal Identifying Information about the data creators or subjects? -->
<!-- scope: telescope -->
no PII
#### Justification for no PII
<!-- info: Provide a justification for selecting `no PII` above. -->
<!-- scope: periscope -->
The SGD dataset does not use identity categories and does not contain sensitive data.
### Maintenance
#### Any Maintenance Plan?
<!-- info: Does the original dataset have a maintenance plan? -->
<!-- scope: telescope -->
no
## Broader Social Context
### Previous Work on the Social Impact of the Dataset
#### Usage of Models based on the Data
<!-- info: Are you aware of cases where models trained on the task featured in this dataset ore related tasks have been used in automated systems? -->
<!-- scope: telescope -->
no
### Impact on Under-Served Communities
#### Addresses needs of underserved Communities?
<!-- info: Does this dataset address the needs of communities that are traditionally underserved in language technology, and particularly language generation technology? Communities may be underserved for exemple because their language, language variety, or social or geographical context is underepresented in NLP and NLG resources (datasets and models). -->
<!-- scope: telescope -->
no
### Discussion of Biases
#### Any Documented Social Biases?
<!-- info: Are there documented social biases in the dataset? Biases in this context are variations in the ways members of different social categories are represented that can have harmful downstream consequences for members of the more disadvantaged group. -->
<!-- scope: telescope -->
no
#### Are the Language Producers Representative of the Language?
<!-- info: Does the distribution of language producers in the dataset accurately represent the full distribution of speakers of the language world-wide? If not, how does it differ? -->
<!-- scope: periscope -->
Due to the combination of the automatic generation and crowd rater paraphasing, the language can be very formulaic. While this may be acceptable for the model part (i.e., we may actually desire an automated agent to form formulaic responses), the input utterances of the simulated customers likely do not cover the entire spectrum of the English language.
## Considerations for Using the Data
### PII Risks and Liability
### Licenses
#### Copyright Restrictions on the Dataset
<!-- info: Based on your answers in the Intended Use part of the Data Overview Section, which of the following best describe the copyright and licensing status of the dataset? -->
<!-- scope: periscope -->
`open license - commercial use allowed`
#### Copyright Restrictions on the Language Data
<!-- info: Based on your answers in the Language part of the Data Curation Section, which of the following best describe the copyright and licensing status of the underlying language data? -->
<!-- scope: periscope -->
`open license - commercial use allowed`
### Known Technical Limitations
#### Technical Limitations
<!-- info: Describe any known technical limitations, such as spurrious correlations, train/test overlap, annotation biases, or mis-annotations, and cite the works that first identified these limitations when possible. -->
<!-- scope: microscope -->
The dialogues under each domain distributed unevenly, where the flights domain has 3644 dialogues while the payment domain only contains 222 dialogues.
Besides, all dialogues are paraphrased by crowd-workers, and it is possible that crow-workers with different culture backgrounds will exhibit biased opinions.
#### Unsuited Applications
<!-- info: When using a model trained on this dataset in a setting where users or the public may interact with its predictions, what are some pitfalls to look out for? In particular, describe some applications of the general task featured in this dataset that its curation or properties make it less suitable for. -->
<!-- scope: microscope -->
Since the initial data was automatically generated, the coverage of entity names is necessarily biased. An agent thus needs to be evaluated in a more realistic environment.
|
true |
# Dataset Card for Arabic Jordanian General Tweets
## Table of Contents
- [Dataset Card for Arabic Jordanian General Tweets](#dataset-card-for-arabic-jordanian-general-tweets)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [|split|num examples|](#splitnum-examples)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Initial Data Collection and Normalization](#initial-data-collection-and-normalization)
- [Who are the source language producers?](#who-are-the-source-language-producers)
- [Annotations](#annotations)
- [Annotation process](#annotation-process)
- [Who are the annotators?](#who-are-the-annotators)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** [Arabic Jordanian General Tweets](https://github.com/komari6/Arabic-twitter-corpus-AJGT)
- **Paper:** [Arabic Tweets Sentimental Analysis Using Machine Learning](https://link.springer.com/chapter/10.1007/978-3-319-60042-0_66)
- **Point of Contact:** [Khaled Alomari](khaled.alomari@adu.ac.ae)
### Dataset Summary
Arabic Jordanian General Tweets (AJGT) Corpus consisted of 1,800 tweets annotated as positive and negative. Modern Standard Arabic (MSA) or Jordanian dialect.
### Supported Tasks and Leaderboards
The dataset was published on this [paper](https://link.springer.com/chapter/10.1007/978-3-319-60042-0_66).
### Languages
The dataset is based on Arabic.
## Dataset Structure
### Data Instances
A binary datset with with negative and positive sentiments.
### Data Fields
- `text` (str): Tweet text.
- `label` (int): Sentiment.
### Data Splits
The dataset is not split.
| | train |
|----------|------:|
| no split | 1,800 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
[More Information Needed]
#### Initial Data Collection and Normalization
Contains 1,800 tweets collected from twitter.
#### Who are the source language producers?
From tweeter.
### Annotations
The dataset does not contain any additional annotations.
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{alomari2017arabic,
title={Arabic tweets sentimental analysis using machine learning},
author={Alomari, Khaled Mohammad and ElSherif, Hatem M and Shaalan, Khaled},
booktitle={International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems},
pages={602--610},
year={2017},
organization={Springer}
}
```
### Contributions
Thanks to [@zaidalyafeai](https://github.com/zaidalyafeai), [@lhoestq](https://github.com/lhoestq) for adding this dataset. |
false |
# Dataset Card for tedlium
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-instances)
- [Data Splits](#data-instances)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** [TED-LIUM homepage](https://www.openslr.org/7/)
- **Repository:** [Needs More Information]
- **Paper:** [TED-LIUM: an Automatic Speech Recognition dedicated corpus](https://aclanthology.org/L12-1405/)
- **Leaderboard:** [Paperswithcode Leaderboard](https://paperswithcode.com/sota/speech-recognition-on-tedlium)
- **Point of Contact:** [Sanchit Gandhi](mailto:sanchit@huggingface.co)
### Dataset Summary
The TED-LIUM corpus is English-language TED talks, with transcriptions, sampled at 16kHz. The three releases of the corpus range from 118 to 452 hours of transcribed speech data.
### Example
```python
from datasets import load_dataset
tedlium = load_dataset("LIUM/tedlium", "release1") # for Release 1
# see structure
print(tedlium)
# load audio sample on the fly
audio_input = tedlium["train"][0]["audio"] # first decoded audio sample
transcription = tedlium["train"][0]["text"] # first transcription
```
### Supported Tasks and Leaderboards
- `automatic-speech-recognition`: The dataset can be used to train a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER). The task has an active leaderboard which can be found at https://paperswithcode.com/sota/speech-recognition-on-tedlium that ranks models based on their WER.
### Languages
The audio and transcriptions are in English, as per the TED talks at http://www.ted.com.
## Dataset Structure
### Data Instances
```
{'audio': {'path': '/home/sanchitgandhi/cache/downloads/extracted/6e3655f9e735ae3c467deed1df788e0dabd671c1f3e2e386e30aa3b571bd9761/TEDLIUM_release1/train/sph/PaulaScher_2008P.sph',
'array': array([-0.00048828, -0.00018311, -0.00137329, ..., 0.00079346,
0.00091553, 0.00085449], dtype=float32),
'sampling_rate': 16000},
'text': '{COUGH} but <sil> i was so {COUGH} utterly unqualified for(2) this project and {NOISE} so utterly ridiculous {SMACK} and ignored the brief {SMACK} <sil>',
'speaker_id': 'PaulaScher_2008P',
'gender': 'female',
'file': '/home/sanchitgandhi/cache/downloads/extracted/6e3655f9e735ae3c467deed1df788e0dabd671c1f3e2e386e30aa3b571bd9761/TEDLIUM_release1/train/sph/PaulaScher_2008P.sph',
'id': 'PaulaScher_2008P-1003.35-1011.16-<o,f0,female>'}
```
### Data Fields
- audio: A dictionary containing the path to the downloaded audio file, the decoded audio array, and the sampling rate. Note that when accessing the audio column: `dataset[0]["audio"]` the audio file is automatically decoded and resampled to `dataset.features["audio"].sampling_rate`. Decoding and resampling of a large number of audio files might take a significant amount of time. Thus it is important to first query the sample index before the `"audio"` column, *i.e.* `dataset[0]["audio"]` should **always** be preferred over `dataset["audio"][0]`.
- file: A path to the downloaded audio file in .sph format.
- text: the transcription of the audio file.
- gender: the gender of the speaker. One of: male, female or N/A.
- id: unique id of the data sample.
- speaker_id: unique id of the speaker. The same speaker id can be found for multiple data samples.
### Data Splits
There are three releases for the TED-LIUM corpus, progressively increasing the number of transcribed speech training data from 118 hours (Release 1), to 207 hours (Release 2), to 452 hours (Release 3).
Release 1:
- 774 audio talks and automatically aligned transcriptions.
- Contains 118 hours of speech audio data.
- Homepage: https://www.openslr.org/7/
Release 2:
- 1495 audio talks and automatically aligned transcriptions.
- Contains 207 hours of speech audio data.
- Dictionary with pronunciations (159848 entries).
- Selected monolingual data for language modeling from WMT12 publicly available corpora.
- Homepage: https://www.openslr.org/19/
Release 3:
- 2351 audio talks and automatically aligned transcriptions.
- Contains 452 hours of speech audio data.
- TED-LIUM 2 validation and test data: 19 TED talks with their corresponding manual transcriptions.
- Dictionary with pronunciations (159848 entries), the same file as the one included in TED-LIUM 2.
- Selected monolingual data for language modeling from WMT12 publicly available corpora: these files come from the TED-LIUM 2 release, but have been modified to produce a tokenization more relevant for English language.
- Homepage: https://www.openslr.org/51/
Release 3 contains two different corpus distributions:
- The ‘legacy’ one, on which the dev and test datasets are the same as in TED-LIUM 2 (and TED-LIUM 1).
- The ‘speaker adaptation’ one, specially designed for experiments on speaker adaptation.
Each release is split into a training, validation and test set:
| Split | Release 1 | Release 2 | Release 3 |
|------------|-----------|-----------|-----------|
| Train | 56,803 | 92,973 | 268,263 |
| Validation | 591 | 591 | 591 |
| Test | 1,469 | 1,469 | 1,469 |
## Dataset Creation
### Curation Rationale
TED-LIUM was built during [The International Workshop on Spoken Language Trans- lation (IWSLT) 2011 Evaluation Campaign](https://aclanthology.org/2011.iwslt-evaluation.1/), an annual workshop focused on the automatic translation of public talks and included tracks for speech recognition, speech translation, text translation, and system combination.
### Source Data
#### Initial Data Collection and Normalization
The data was obtained from publicly available TED talks at http://www.ted.com. Proper alignments between the speech and the transcribed text were generated using an in-house speaker segmentation and clustering tool (_LIUM_SpkDiarization_). Speech disfluencies (e.g. repetitions, hesitations, false starts) were treated in the following way: repetitions were transcribed, hesitations mapped to a specific filler word, and false starts not taken into account. For full details on the data collection and processing, refer to the [TED-LIUM paper](https://aclanthology.org/L12-1405/).
#### Who are the source language producers?
TED Talks are influential videos from expert speakers on education, business, science, tech and creativity.
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
Licensed under Creative Commons BY-NC-ND 3.0 (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en).
### Citation Information
Release 1:
```
@inproceedings{rousseau2012tedlium,
title={TED-LIUM: an Automatic Speech Recognition dedicated corpus},
author={Rousseau, Anthony and Del{\'e}glise, Paul and Est{\`e}ve, Yannick},
booktitle={Conference on Language Resources and Evaluation (LREC)},
pages={125--129},
year={2012}
}
```
Release 2:
```
@inproceedings{rousseau2014enhancing,
title={Enhancing the TED-LIUM corpus with selected data for language modeling and more TED talks.},
author={Rousseau, Anthony and Del{\'e}glise, Paul and Esteve, Yannick and others},
booktitle={LREC},
pages={3935--3939},
year={2014}
}
```
Release 3:
```
@inproceedings{hernandez2018ted,
author="Hernandez, Fran{\c{c}}ois
and Nguyen, Vincent
and Ghannay, Sahar
and Tomashenko, Natalia
and Est{\`e}ve, Yannick",
title="TED-LIUM 3: Twice as Much Data and Corpus Repartition for Experiments on Speaker Adaptation",
booktitle="Speech and Computer",
year="2018",
publisher="Springer International Publishing",
pages="198--208",
}
``` |
false |
# Dataset Card for COVID-QA
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** https://github.com/deepset-ai/COVID-QA
- **Paper:** https://openreview.net/forum?id=JENSKEEzsoU
- **Point of Contact:** [deepset AI](https://github.com/deepset-ai)
### Dataset Summary
COVID-QA is a Question Answering dataset consisting of 2,019 question/answer pairs annotated by volunteer biomedical experts on scientific articles related to COVID-19.
A total of 147 scientific articles from the CORD-19 dataset were annotated by 15 experts.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
The text in the dataset is in English.
## Dataset Structure
### Data Instances
**What do the instances that comprise the dataset represent?**
Each represents a question, a context (document passage from the CORD19 dataset) and an answer.
**How many instances are there in total?**
2019 instances
**What data does each instance consist of?**
Each instance is a question, a set of answers, and an id associated with each answer.
[More Information Needed]
### Data Fields
The data was annotated in SQuAD style fashion, where each row contains:
* **question**: Query question
* **context**: Context text to obtain the answer from
* **document_id** The document ID of the context text
* **answer**: Dictionary containing the answer string and the start index
### Data Splits
**data/COVID-QA.json**: 2,019 question/answer pairs annotated by volunteer biomedical experts on scientific articles related to COVID-19.
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
The inital data collected comes from 147 scientific articles from the CORD-19 dataset. Question and answers were then
annotated afterwards.
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
While annotators were volunteers, they were required to have at least a Master’s degree in biomedical sciences.
The annotation team was led by a medical doctor (G.A.R.) who vetted the volunteer’s credentials and
manually verified each question/answer pair produced. We used an existing, web-based annotation tool that had been
created by deepset and is available at their Neural Search framework [haystack](https://github.com/deepset-ai/haystack).
#### Who are the annotators?
The annotators are 15 volunteer biomedical experts on scientific articles related to COVID-19.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
The dataset aims to help build question answering models serving clinical and scientific researchers, public health authorities, and frontline workers.
These QA systems can help them find answers and patterns in research papers by locating relevant answers to common questions from scientific articles.
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
## Additional Information
The listed authors in the homepage are maintaining/supporting the dataset.
### Dataset Curators
[More Information Needed]
### Licensing Information
The Proto_qa dataset is licensed under the [Apache License 2.0](https://github.com/deepset-ai/COVID-QA/blob/master/LICENSE)
### Citation Information
```
@inproceedings{moller2020covid,
title={COVID-QA: A Question Answering Dataset for COVID-19},
author={M{\"o}ller, Timo and Reina, Anthony and Jayakumar, Raghavan and Pietsch, Malte},
booktitle={Proceedings of the 1st Workshop on NLP for COVID-19 at ACL 2020},
year={2020}
}
```
### Contributions
Thanks to [@olinguyen](https://github.com/olinguyen) for adding this dataset. |
false |
# Dataset Card Creation Guide
## Table of Contents
- [Dataset Card Creation Guide](#dataset-card-creation-guide)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Initial Data Collection and Normalization](#initial-data-collection-and-normalization)
- [Who are the source language producers?](#who-are-the-source-language-producers)s
- [Additional Information](#additional-information)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [stackexchange](https://archive.org/details/stackexchange)
- **Repository:** [flax-sentence-embeddings](https://github.com/nreimers/flax-sentence-embeddings)
### Dataset Summary
We automatically extracted question and answer (Q&A) pairs from [Stack Exchange](https://stackexchange.com/) network. Stack Exchange gather many Q&A communities across 50 online plateform, including the well known Stack Overflow and other technical sites. 100 millon developpers consult Stack Exchange every month. The dataset is a parallel corpus with each question mapped to the top rated answer. The dataset is split given communities which cover a variety of domains from 3d printing, economics, raspberry pi or emacs. An exhaustive list of all communities is available [here](https://stackexchange.com/sites).
### Languages
Stack Exchange mainly consist of english language (en).
## Dataset Structure
### Data Instances
Each data samples is presented as follow:
```
{'title_body': 'How to determine if 3 points on a 3-D graph are collinear? Let the points $A, B$ and $C$ be $(x_1, y_1, z_1), (x_2, y_2, z_2)$ and $(x_3, y_3, z_3)$ respectively. How do I prove that the 3 points are collinear? What is the formula?',
'upvoted_answer': 'From $A(x_1,y_1,z_1),B(x_2,y_2,z_2),C(x_3,y_3,z_3)$ we can get their position vectors.\n\n$\\vec{AB}=(x_2-x_1,y_2-y_1,z_2-z_1)$ and $\\vec{AC}=(x_3-x_1,y_3-y_1,z_3-z_1)$.\n\nThen $||\\vec{AB}\\times\\vec{AC}||=0\\implies A,B,C$ collinear.',
'downvoted_answer': 'If the distance between |AB|+|BC|=|AC| then A,B,C are collinear.'}
```
This particular exampe corresponds to the [following page](https://math.stackexchange.com/questions/947555/how-to-determine-if-3-points-on-a-3-d-graph-are-collinear)
### Data Fields
The fields present in the dataset contain the following informations:
- `title_body`: This is the concatenation of the title and body from the question
- `upvoted_answer`: This is the body from the most upvoted answer
- `downvoted_answer`: This is the body from most downvoted answer
- `title`: This is the title from the question
### Data Splits
We provide three splits for this dataset, which only differs by the structure of the fieds which are retrieved:
- `titlebody_upvoted_downvoted_answer`: Includes title and body from the question as well as most upvoted and downvoted answer.
- `title_answer`: Includes title from the question as well as most upvoted answer.
- `titlebody_answer`: Includes title and body from the question as well as most upvoted answer.
| | Number of pairs |
| ----- | ------ |
| `titlebody_upvoted_downvoted_answer` | 17,083 |
| `title_answer` | 1,100,953 |
| `titlebody_answer` | 1,100,953 |
## Dataset Creation
### Curation Rationale
We primary designed this dataset for sentence embeddings training. Indeed sentence embeddings may be trained using a contrastive learning setup for which the model is trained to associate each sentence with its corresponding pair out of multiple proposition. Such models require many examples to be efficient and thus the dataset creation may be tedious. Community networks such as Stack Exchange allow us to build many examples semi-automatically.
### Source Data
The source data are dumps from [Stack Exchange](https://archive.org/details/stackexchange)
#### Initial Data Collection and Normalization
We collected the data from the math community.
We filtered out questions which title or body length is bellow 20 characters and questions for which body length is above 4096 characters.
When extracting most upvoted answer, we filtered to pairs for which their is at least 100 votes gap between most upvoted and downvoted answers.
#### Who are the source language producers?
Questions and answers are written by the community developpers of Stack Exchange.
## Additional Information
### Licensing Information
Please see the license information at: https://archive.org/details/stackexchange
### Citation Information
```
@misc{StackExchangeDataset,
author = {Flax Sentence Embeddings Team},
title = {Stack Exchange question pairs},
year = {2021},
howpublished = {https://huggingface.co/datasets/flax-sentence-embeddings/},
}
```
### Contributions
Thanks to the Flax Sentence Embeddings team for adding this dataset. |
false |
# Dataset Card for QUAERO
## Dataset Description
- **Homepage:** https://quaerofrenchmed.limsi.fr/
- **Pubmed:** True
- **Public:** True
- **Tasks:** Named-Entity Recognition (NER)
The QUAERO French Medical Corpus has been initially developed as a resource for named entity recognition and normalization [1]. It was then improved with the purpose of creating a gold standard set of normalized entities for French biomedical text, that was used in the CLEF eHealth evaluation lab [2][3].
A selection of MEDLINE titles and EMEA documents were manually annotated. The annotation process was guided by concepts in the Unified Medical Language System (UMLS):
1. Ten types of clinical entities, as defined by the following UMLS Semantic Groups (Bodenreider and McCray 2003) were annotated: Anatomy, Chemical and Drugs, Devices, Disorders, Geographic Areas, Living Beings, Objects, Phenomena, Physiology, Procedures.
2. The annotations were made in a comprehensive fashion, so that nested entities were marked, and entities could be mapped to more than one UMLS concept. In particular: (a) If a mention can refer to more than one Semantic Group, all the relevant Semantic Groups should be annotated. For instance, the mention “récidive” (recurrence) in the phrase “prévention des récidives” (recurrence prevention) should be annotated with the category “DISORDER” (CUI C2825055) and the category “PHENOMENON” (CUI C0034897); (b) If a mention can refer to more than one UMLS concept within the same Semantic Group, all the relevant concepts should be annotated. For instance, the mention “maniaques” (obsessive) in the phrase “patients maniaques” (obsessive patients) should be annotated with CUIs C0564408 and C0338831 (category “DISORDER”); (c) Entities which span overlaps with that of another entity should still be annotated. For instance, in the phrase “infarctus du myocarde” (myocardial infarction), the mention “myocarde” (myocardium) should be annotated with category “ANATOMY” (CUI C0027061) and the mention “infarctus du myocarde” should be annotated with category “DISORDER” (CUI C0027051)
The QUAERO French Medical Corpus BioC release comprises a subset of the QUAERO French Medical corpus, as follows:
Training data (BRAT version used in CLEF eHealth 2015 task 1b as training data):
- MEDLINE_train_bioc file: 833 MEDLINE titles, annotated with normalized entities in the BioC format
- EMEA_train_bioc file: 3 EMEA documents, segmented into 11 sub-documents, annotated with normalized entities in the BioC format
Development data (BRAT version used in CLEF eHealth 2015 task 1b as test data and in CLEF eHealth 2016 task 2 as development data):
- MEDLINE_dev_bioc file: 832 MEDLINE titles, annotated with normalized entities in the BioC format
- EMEA_dev_bioc file: 3 EMEA documents, segmented into 12 sub-documents, annotated with normalized entities in the BioC format
Test data (BRAT version used in CLEF eHealth 2016 task 2 as test data):
- MEDLINE_test_bioc folder: 833 MEDLINE titles, annotated with normalized entities in the BioC format
- EMEA folder_test_bioc: 4 EMEA documents, segmented into 15 sub-documents, annotated with normalized entities in the BioC format
This release of the QUAERO French medical corpus, BioC version, comes in the BioC format, through automatic conversion from the original BRAT format obtained with the Brat2BioC tool https://bitbucket.org/nicta_biomed/brat2bioc developped by Jimeno Yepes et al.
Antonio Jimeno Yepes, Mariana Neves, Karin Verspoor
Brat2BioC: conversion tool between brat and BioC
BioCreative IV track 1 - BioC: The BioCreative Interoperability Initiative, 2013
Please note that the original version of the QUAERO corpus distributed in the CLEF eHealth challenge 2015 and 2016 came in the BRAT stand alone format. It was distributed with the CLEF eHealth evaluation tool. This original distribution of the QUAERO French Medical corpus is available separately from https://quaerofrenchmed.limsi.fr
All questions regarding the task or data should be addressed to aurelie.neveol@limsi.fr
## Citation Information
```
@InProceedings{neveol14quaero,
author = {Névéol, Aurélie and Grouin, Cyril and Leixa, Jeremy
and Rosset, Sophie and Zweigenbaum, Pierre},
title = {The {QUAERO} {French} Medical Corpus: A Ressource for
Medical Entity Recognition and Normalization},
OPTbooktitle = {Proceedings of the Fourth Workshop on Building
and Evaluating Ressources for Health and Biomedical
Text Processing},
booktitle = {Proc of BioTextMining Work},
OPTseries = {BioTxtM 2014},
year = {2014},
pages = {24--30},
}
``` |
false |
# Dataset Card for [Dataset Name]
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://ec.europa.eu/jrc/en/language-technologies/ecdc-translation-memory](https://ec.europa.eu/jrc/en/language-technologies/ecdc-translation-memory)
- **Paper:** [https://link.springer.com/article/10.1007/s10579-014-9277-0](https://link.springer.com/article/10.1007/s10579-014-9277-0)
- **Point of Contact:** [Ralf Steinberger](mailto:Ralf.Steinberger@jrc.ec.europa.eu)
### Dataset Summary
In October 2012, the European Union (EU) agency 'European Centre for Disease Prevention and Control' (ECDC) released a translation memory (TM), i.e. a collection of sentences and their professionally produced translations, in twenty-five languages.
ECDC-TM covers 25 languages: the 23 official languages of the EU plus Norwegian (Norsk) and Icelandic. ECDC-TM was created by translating from English into the following 24 languages: Bulgarian, Czech, Danish, Dutch, English, Estonian, Gaelige (Irish), German, Greek, Finnish, French, Hungarian, Icelandic, Italian, Latvian, Lithuanian, Maltese, Norwegian (NOrsk), Polish, Portuguese, Romanian, Slovak, Slovenian, Spanish and Swedish.
All documents and sentences were originally written in English. They were then translated into the other languages by professional translators from the Translation Centre CdT in Luxembourg.
To load a language pair that is not part of the config, just specify the language code as language pair. For example, if you want to translate Czech to Greek:
`dataset = load_dataset("europa_ecdc_tm", language_pair=("cs", "el"))`
### Supported Tasks and Leaderboards
- `text2text-generation`: the dataset can be used to train a model for `machine-translation`. Machine translation models are usually evaluated using metrics such as [BLEU](https://huggingface.co/metrics/bleu), [ROUGE](https://huggingface.co/metrics/rouge) or [SacreBLEU](https://huggingface.co/metrics/sacrebleu). You can use the [mBART](https://huggingface.co/facebook/mbart-large-cc25) model for this task. This task has active leaderboards which can be found at [https://paperswithcode.com/task/machine-translation](https://paperswithcode.com/task/machine-translation), which usually rank models based on [BLEU score](https://huggingface.co/metrics/bleu).
### Languages
All documents and sentences were originally written in English (`en`). They were then translated into the other languages by professional translators from the Translation Centre CdT in Luxembourg.
Translations are available in these languages: `en`, `bg`, `cs`, `da`, `de`, `el`, `en`, `es`, `et`, `fi`, `fr`, `ga`, `hu`, `is`, `it`, `lt`, `lv`, `mt`, `nl`, `no`, `pl`, `pt`, `ro`, `sk`, `sl`, `sv`.
## Dataset Structure
### Data Instances
```
{
"translation": {
"<source_language>":"Sentence to translate",
"<target_language>": "Translated sentence",
},
}
```
### Data Fields
- `translation`: a multilingual `string` variable, with possible languages including `en`, `bg`, `cs`, `da`, `de`, `el`, `en`, `es`, `et`, `fi`, `fr`, `ga`, `hu`, `is`, `it`, `lt`, `lv`, `mt`, `nl`, `no`, `pl`, `pt`, `ro`, `sk`, `sl`, `sv`.
### Data Splits
The data is not splitted (only the `train` split is available).
## Dataset Creation
### Curation Rationale
The ECDC-TM is relatively small compared to the JRC-Acquis and to DGT-TM, but it has the advantage that it focuses on a very different domain, namely that of public health. Also, it includes translation units for the languages Irish (Gaelige, GA), Norwegian (Norsk, NO) and Icelandic (IS).
### Source Data
#### Initial Data Collection and Normalization
ECDC-TM was built on the basis of the website of the European Centre for Disease Prevention and Control (ECDC). The major part of the documents talks about health-related topics (anthrax, botulism, cholera, dengue fever, hepatitis, etc.), but some of the web pages also describe the organisation ECDC (e.g. its organisation, job opportunities) and its activities (e.g. epidemic intelligence, surveillance).
#### Who are the source language producers?
All documents and sentences were originally written in English, by the ECDC website content producers.
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
All documents and sentences were thus originally written in English. They were then translated into the other languages by professional translators from the Translation Centre CdT in Luxembourg.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
Contains translations of sentences in the public healthcare domain, including technical terms (disease and treatment names for example).
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
Copyright © EU / ECDC, 2020
#### Copyright
The Work (as defined below) is provided under the terms of this Licence (or later versions of
this Licence published by the European Commission). The work is protected by copyright
and/or other applicable law. Any use of the work other than as authorised under this
Licence or copyright law is prohibited.
The terms provided herein conform to the reuse policy established by the Commission's
Reuse Decision (2011/833/EU).
By exercising any rights to the work provided here, you accept and agree to be bound by the
terms of this Licence. The Owner (as defined below) grants You the rights conferred by this
Licence in consideration of your acceptance of such terms and conditions.
#### Definitions
The ‘Owner’ shall mean jointly the European Union represented by the European
Commission and the European Centre for Disease Prevention and Control, which are the
original licensors and/or control the copyright and any other intellectual and industrial
property rights related to the Work.
The ‘Work’ is the information and/or data offered to You under this Licence, according to
the ‘Copyright Notice’:
Copyright (c) EU/ECDC, <YEAR>
‘You’ means the natural or legal person, or body of persons corporate or incorporate,
acquiring rights under this Licence.
‘Use’ means any act which is restricted by copyright or database rights, whether in the
original medium or in any other medium, and includes, without limitation, distributing,
copying, adapting, or modifying as may be technically necessary to use the Work in a
different mode or format. It includes ‘re‐Use’, meaning the use, communication to the
public and/or distribution of the Works for purposes other than the initial purpose for which
the Work was produced.
#### Rights
You are herewith granted a worldwide, royalty‐free, perpetual, non‐exclusive Licence to Use
and re‐Use the Works and any modifications thereof for any commercial and non‐
commercial purpose allowed by the law, provided that the following conditions are met:
a) Unmodified distributions must retain the above Copyright Notice;
b) Unmodified distributions must retain the following ‘No Warranty’ disclaimer;
c) You will not use the name of the Owner to endorse or promote products and
services derived from Use of the Work without specific prior written permission.
#### No warranty
Each Work is provided ‘as is’ without, to the full extent permitted by law, representations,
warranties, obligations and liabilities of any kind, either express or implied, including, but
not limited to, any implied warranty of merchantability, integration, satisfactory quality and
fitness for a particular purpose.
Except in the cases of wilful misconduct or damages directly caused to natural persons, the
Owner will not be liable for any incidental, consequential, direct or indirect damages,
including, but not limited to, the loss of data, lost profits or any other financial loss arising
from the use of, or inability to use, the Work even if the Owner has been notified of the
possibility of such loss, damages, claims or costs, or for any claim by any third party. The
Owner may be liable under national statutory product liability laws as far as such laws apply
to the Work.
### Citation Information
```
@Article{Steinberger2014,
author={Steinberger, Ralf
and Ebrahim, Mohamed
and Poulis, Alexandros
and Carrasco-Benitez, Manuel
and Schl{\"u}ter, Patrick
and Przybyszewski, Marek
and Gilbro, Signe},
title={An overview of the European Union's highly multilingual parallel corpora},
journal={Language Resources and Evaluation},
year={2014},
month={Dec},
day={01},
volume={48},
number={4},
pages={679-707},
issn={1574-0218},
doi={10.1007/s10579-014-9277-0},
url={https://doi.org/10.1007/s10579-014-9277-0}
}
```
### Contributions
Thanks to [@SBrandeis](https://github.com/SBrandeis) for adding this dataset. |
false |
# Dataset Description
The `proof-pile` is a 13GB pre-training dataset of mathematical text that comprises 8.3 billion tokens (using the `gpt-neox` tokenizer). Models trained on this dataset are coming soon :) The dataset is composed of diverse sources of both informal and formal mathematics, namely
- ArXiv.math (10GB)
- Open-source math textbooks (50MB)
- Formal mathematics libraries (500MB)
- Lean mathlib and other Lean repositories
- Isabelle AFP
- Coq mathematical components and other Coq repositories
- HOL Light
- set.mm
- Mizar Mathematical Library
- Math Overflow and Math Stack Exchange (2.5GB)
- Wiki-style sources (50MB)
- ProofWiki
- Wikipedia math articles
- MATH dataset (6MB)
The construction of the dataset is reproducible using the code and instructions in the [proof-pile Github
repo](https://github.com/zhangir-azerbayev/proof-pile).
# Supported Tasks
This dataset is intended to be used for pre-training and fine-tuning language models. We envision models trained on the `proof-pile` will have many downstream applications, including informal quantitative reasoning, formal theorem proving, semantic search for formal mathematics, and autoformalization.
# Languages
All informal mathematics in the `proof-pile` is written in English and LaTeX (arXiv articles in other languages are filtered out using [languagedetect](https://github.com/shuyo/language-detection/blob/wiki/ProjectHome.md)). Formal theorem proving languages represented in this dataset are Lean 3, Isabelle, Coq, HOL Light, Metamath, and Mizar.
# Evaluation
The version of `set.mm` in this dataset has 10% of proofs replaced with the `?` character in order to preserve a validation and test set for Metamath provers pre-trained on the `proof-pile`. The precise split can be found here: [validation](https://github.com/zhangir-azerbayev/mm-extract/blob/main/valid_decls.json) and [test](https://github.com/zhangir-azerbayev/mm-extract/blob/main/test_decls.json).
The Lean mathlib commit used in this dataset is `6313863`. Theorems created in subsequent commits can be used for evaluating Lean theorem provers.
This dataset contains only the training set of the [MATH dataset](https://github.com/hendrycks/math). However, because this dataset contains ProofWiki, the Stacks Project, Trench's Analysis, and Stein's Number Theory, models trained on it cannot be evaluated on the [NaturalProofs dataset](https://github.com/wellecks/naturalproofs).
# Data Preprocessing
This section describes any significant filtering and transformations made to various subsets of the data.
**arXiv.math.**
The arXiv.math dataset is large, heterogeneous, and contains a great deal of noise. We used the following heuristics
when choosing which files from arXiv.math source folders to include in the dataset:
- Keep only files with a `.tex` extension.
- Only include files that use either a `utf-8/16/32` or `latin-1` text encoding.
- Discard files that do not contain a part, chapter, section, sub...section, paragraph, or subparagraph heading.
- Delete files that contain the keyword `gnuplot`. Gnuplot-latex is an old command line utility that generates blocks
of entirely unintelligible source.
- Include only articles in English, as determined by the [langdetect library](https://pypi.org/project/langdetect/). \n",
"\n",
- Exclude files shorter than 280 characters (characters counted after substring removal described below).
In addition, we apply the following transformations to arXiv.math texts:
- Delete everything outside of `\begin{document}` and `\end{document}`.
- Delete everything including or after `\Refs`, `\begin{thebibliography}`, or `\begin{bibdiv}`
- Delete comments.
- Any more than three consecutive newlines are replaced by three consecutive newlines.
In [this notebook](https://github.com/zhangir-azerbayev/proof-pile/blob/main/analysis/arxiv_noisedetection.ipynb), we provide an analysis of the prevalence of noisy documents in the arXiv.math subset of the
proof-pile.
**Stack Exchange.**
We only include questions that have at least 5 upvotes and an answer. We format Stack Exchange posts as follows
```
QUESTION [{num_upvotes} upvotes]: {text of question}
REPLY [{num_upvotes} votes]: {text of reply}
REPLY [{num_upvotes} votes]: {text of reply}
.
.
.
```
**set.mm.**
We converted `set.mm` into human-readable form by following the instructions in the [mm-extract repo](https://github.com/zhangir-azerbayev/mm-extract)
## Contributions
Authors: Zhangir Azerbayev, Edward Ayers, Bartosz Piotrowski.
We would like to thank Jeremy Avigad, Albert Jiang, and Wenda Li for their invaluable guidance, and the Hoskinson Center for Formal Mathematics for its support.
|
true |
# Dataset Card for WRIME
[](https://github.com/shunk031/huggingface-datasets_wrime/actions/workflows/ci.yaml)
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- Homepage: https://github.com/ids-cv/wrime
- Repository: https://github.com/shunk031/huggingface-datasets_wrime
- Paper: https://aclanthology.org/2021.naacl-main.169/
### Dataset Summary
In this study, we introduce a new dataset, WRIME, for emotional intensity estimation. We collect both the subjective emotional intensity ofthe writers themselves and the objective one annotated by the readers, and explore the differences between them. In our data collection, we hired 50 participants via crowdsourcing service. They annotated their own past posts on a social networking service (SNS) with the subjective emotional intensity. We also hired 3 annotators, who annotated allposts with the objective emotional intensity. Consequently, our Japanese emotion analysis datasetconsists of 17,000 posts with both subjective andobjective emotional intensities for Plutchik’s eightemotions ([Plutchik, 1980](https://www.sciencedirect.com/science/article/pii/B9780125587013500077)), which are given in afour-point scale (no, weak, medium, and strong).
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
- Japanese
## Dataset Structure
### Data Instances
When loading a specific configuration, users has to append a version dependent suffix:
```python
from datasets import load_dataset
dataset = load_dataset("shunk031/wrime", name="ver1")
print(dataset)
# DatasetDict({
# train: Dataset({
# features: ['sentence', 'user_id', 'datetime', 'writer', 'reader1', 'reader2', 'reader3', 'avg_readers'],
# num_rows: 40000
# })
# validation: Dataset({
# features: ['sentence', 'user_id', 'datetime', 'writer', 'reader1', 'reader2', 'reader3', 'avg_readers'],
# num_rows: 1200
# })
# test: Dataset({
# features: ['sentence', 'user_id', 'datetime', 'writer', 'reader1', 'reader2', 'reader3', 'avg_readers'],
# num_rows: 2000
# })
# })
```
#### Ver. 1
An example of looks as follows:
```json
{
"sentence": "ぼけっとしてたらこんな時間｡チャリあるから食べにでたいのに…",
"user_id": "1",
"datetime": "2012/07/31 23:48",
"writer": {
"joy": 0,
"sadness": 1,
"anticipation": 2,
"surprise": 1,
"anger": 1,
"fear": 0,
"disgust": 0,
"trust": 1
},
"reader1": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 0,
"anger": 0,
"fear": 0,
"disgust": 0,
"trust": 0
},
"reader2": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 1,
"anger": 0,
"fear": 0,
"disgust": 0,
"trust": 0
},
"reader3": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 0,
"anger": 0,
"fear": 1,
"disgust": 1,
"trust": 0
},
"avg_readers": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 0,
"anger": 0,
"fear": 0,
"disgust": 0,
"trust": 0
}
}
```
#### Ver. 1
An example of looks as follows:
```json
{
"sentence": "ぼけっとしてたらこんな時間。チャリあるから食べにでたいのに…",
"user_id": "1",
"datetime": "2012/7/31 23:48",
"writer": {
"joy": 0,
"sadness": 1,
"anticipation": 2,
"surprise": 1,
"anger": 1,
"fear": 0,
"disgust": 0,
"trust": 1,
"sentiment": 0
},
"reader1": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 0,
"anger": 0,
"fear": 0,
"disgust": 0,
"trust": 0,
"sentiment": -2
},
"reader2": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 0,
"anger": 0,
"fear": 1,
"disgust": 1,
"trust": 0,
"sentiment": -1
},
"reader3": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 1,
"anger": 0,
"fear": 0,
"disgust": 0,
"trust": 0,
"sentiment": -1
},
"avg_readers": {
"joy": 0,
"sadness": 2,
"anticipation": 0,
"surprise": 0,
"anger": 0,
"fear": 0,
"disgust": 0,
"trust": 0,
"sentiment": -1
}
}
```
### Data Fields
#### Ver. 1
- `sentence`: 投稿テキスト
- `user_id`: ユーザー ID
- `datetime`: 投稿日時
- `writer`: 主観 （書き手）
- `joy`: 主観の喜びの感情
- `sadness`: 主観の悲しみの感情
- `anticipation`: 主観の期待の感情
- `surprise`: 主観の驚きの感情
- `anger`: 主観の怒りの感情
- `fear`: 主観の恐れの感情
- `disgust`: 主観の嫌悪の感情
- `trust`: 主観の信頼の感情
- `reader1`: 客観 A （読み手 A)
- `joy`: 客観 A の喜びの感情
- `sadness`: 客観 A の悲しみの感情
- `anticipation`: 客観 A の期待の感情
- `surprise`: 客観 A の驚きの感情
- `anger`: 客観 A の怒りの感情
- `fear`: 客観 A の恐れの感情
- `disgust`: 客観 A の嫌悪の感情
- `trust`: 客観 A の信頼の感情
- `reader2`: 客観 B （読み手 B)
- `joy`: 客観 B の喜びの感情
- `sadness`: 客観 B の悲しみの感情
- `anticipation`: 客観 B の期待の感情
- `surprise`: 客観 B の驚きの感情
- `anger`: 客観 B の怒りの感情
- `fear`: 客観 B の恐れの感情
- `disgust`: 客観 B の嫌悪の感情
- `trust`: 客観 B の信頼の感情
- `reader3`: 客観 C （読み手 C)
- `joy`: 客観 C の喜びの感情
- `sadness`: 客観 C の悲しみの感情
- `anticipation`: 客観 C の期待の感情
- `surprise`: 客観 C の驚きの感情
- `anger`: 客観 C の怒りの感情
- `fear`: 客観 C の恐れの感情
- `disgust`: 客観 C の嫌悪の感情
- `trust`: 客観 C の信頼の感情
- `avg_readers`
- `joy`: 客観 A, B, C 平均の喜びの感情
- `sadness`: 客観 A, B, C 平均の悲しみの感情
- `anticipation`: 客観 A, B, C 平均の期待の感情
- `surprise`: 客観 A, B, C 平均の驚きの感情
- `anger`: 客観 A, B, C 平均の怒りの感情
- `fear`: 客観 A, B, C 平均の恐れの感情
- `disgust`: 客観 A, B, C 平均の嫌悪の感情
- `trust`: 客観 A, B, C 平均の信頼の感情
#### Ver. 2
- `sentence`: 投稿テキスト
- `user_id`: ユーザー ID
- `datetime`: 投稿日時
- `writer`: 主観 （書き手）
- `joy`: 主観の喜びの感情
- `sadness`: 主観の悲しみの感情
- `anticipation`: 主観の期待の感情
- `surprise`: 主観の驚きの感情
- `anger`: 主観の怒りの感情
- `fear`: 主観の恐れの感情
- `disgust`: 主観の嫌悪の感情
- `trust`: 主観の信頼の感情
- `sentiment`: 主観の感情極性
- `reader1`: 客観 A （読み手 A)
- `joy`: 客観 A の喜びの感情
- `sadness`: 客観 A の悲しみの感情
- `anticipation`: 客観 A の期待の感情
- `surprise`: 客観 A の驚きの感情
- `anger`: 客観 A の怒りの感情
- `fear`: 客観 A の恐れの感情
- `disgust`: 客観 A の嫌悪の感情
- `trust`: 客観 A の信頼の感情
- `sentiment`: 客観 A の感情極性
- `reader2`: 客観 B （読み手 B)
- `joy`: 客観 B の喜びの感情
- `sadness`: 客観 B の悲しみの感情
- `anticipation`: 客観 B の期待の感情
- `surprise`: 客観 B の驚きの感情
- `anger`: 客観 B の怒りの感情
- `fear`: 客観 B の恐れの感情
- `disgust`: 客観 B の嫌悪の感情
- `trust`: 客観 B の信頼の感情
- `sentiment`: 客観 B の感情極性
- `reader3`: 客観 C （読み手 C)
- `joy`: 客観 C の喜びの感情
- `sadness`: 客観 C の悲しみの感情
- `anticipation`: 客観 C の期待の感情
- `surprise`: 客観 C の驚きの感情
- `anger`: 客観 C の怒りの感情
- `fear`: 客観 C の恐れの感情
- `disgust`: 客観 C の嫌悪の感情
- `trust`: 客観 C の信頼の感情
- `sentiment`: 客観 C の感情極性
- `avg_readers`
- `joy`: 客観 A, B, C 平均の喜びの感情
- `sadness`: 客観 A, B, C 平均の悲しみの感情
- `anticipation`: 客観 A, B, C 平均の期待の感情
- `surprise`: 客観 A, B, C 平均の驚きの感情
- `anger`: 客観 A, B, C 平均の怒りの感情
- `fear`: 客観 A, B, C 平均の恐れの感情
- `disgust`: 客観 A, B, C 平均の嫌悪の感情
- `trust`: 客観 A, B, C 平均の信頼の感情
- `sentiment`: 客観 A, B, C 平均の感情極性
### Data Splits
| name | train | validation | test |
|------|-------:|-----------:|------:|
| ver1 | 40,000 | 1,200 | 2,000 |
| ver2 | 30,000 | 2,500 | 2,500 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
From [the README](https://github.com/ids-cv/wrime/blob/master/README.en.md#licence) of the GitHub:
- The dataset is available for research purposes only.
- Redistribution of the dataset is prohibited.
### Citation Information
```bibtex
@inproceedings{kajiwara-etal-2021-wrime,
title = "{WRIME}: A New Dataset for Emotional Intensity Estimation with Subjective and Objective Annotations",
author = "Kajiwara, Tomoyuki and
Chu, Chenhui and
Takemura, Noriko and
Nakashima, Yuta and
Nagahara, Hajime",
booktitle = "Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jun,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.naacl-main.169",
doi = "10.18653/v1/2021.naacl-main.169",
pages = "2095--2104",
abstract = "We annotate 17,000 SNS posts with both the writer{'}s subjective emotional intensity and the reader{'}s objective one to construct a Japanese emotion analysis dataset. In this study, we explore the difference between the emotional intensity of the writer and that of the readers with this dataset. We found that the reader cannot fully detect the emotions of the writer, especially anger and trust. In addition, experimental results in estimating the emotional intensity show that it is more difficult to estimate the writer{'}s subjective labels than the readers{'}. The large gap between the subjective and objective emotions imply the complexity of the mapping from a post to the subjective emotion intensities, which also leads to a lower performance with machine learning models.",
}
```
```bibtex
@inproceedings{suzuki-etal-2022-japanese,
title = "A {J}apanese Dataset for Subjective and Objective Sentiment Polarity Classification in Micro Blog Domain",
author = "Suzuki, Haruya and
Miyauchi, Yuto and
Akiyama, Kazuki and
Kajiwara, Tomoyuki and
Ninomiya, Takashi and
Takemura, Noriko and
Nakashima, Yuta and
Nagahara, Hajime",
booktitle = "Proceedings of the Thirteenth Language Resources and Evaluation Conference",
month = jun,
year = "2022",
address = "Marseille, France",
publisher = "European Language Resources Association",
url = "https://aclanthology.org/2022.lrec-1.759",
pages = "7022--7028",
abstract = "We annotate 35,000 SNS posts with both the writer{'}s subjective sentiment polarity labels and the reader{'}s objective ones to construct a Japanese sentiment analysis dataset. Our dataset includes intensity labels (\textit{none}, \textit{weak}, \textit{medium}, and \textit{strong}) for each of the eight basic emotions by Plutchik (\textit{joy}, \textit{sadness}, \textit{anticipation}, \textit{surprise}, \textit{anger}, \textit{fear}, \textit{disgust}, and \textit{trust}) as well as sentiment polarity labels (\textit{strong positive}, \textit{positive}, \textit{neutral}, \textit{negative}, and \textit{strong negative}). Previous studies on emotion analysis have studied the analysis of basic emotions and sentiment polarity independently. In other words, there are few corpora that are annotated with both basic emotions and sentiment polarity. Our dataset is the first large-scale corpus to annotate both of these emotion labels, and from both the writer{'}s and reader{'}s perspectives. In this paper, we analyze the relationship between basic emotion intensity and sentiment polarity on our dataset and report the results of benchmarking sentiment polarity classification.",
}
```
### Contributions
Thanks to [@moguranosenshi](https://github.com/moguranosenshi) for creating this dataset.
|
false |
# Dataset Card for NIH Chest X-ray dataset
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [NIH Chest X-ray Dataset of 10 Common Thorax Disease Categories](https://nihcc.app.box.com/v/ChestXray-NIHCC/folder/36938765345)
- **Repository:**
- **Paper:** [ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases](https://arxiv.org/abs/1705.02315)
- **Leaderboard:**
- **Point of Contact:** rms@nih.gov
### Dataset Summary
_ChestX-ray dataset comprises 112,120 frontal-view X-ray images of 30,805 unique patients with the text-mined fourteen disease image labels (where each image can have multi-labels), mined from the associated radiological reports using natural language processing. Fourteen common thoracic pathologies include Atelectasis, Consolidation, Infiltration, Pneumothorax, Edema, Emphysema, Fibrosis, Effusion, Pneumonia, Pleural_thickening, Cardiomegaly, Nodule, Mass and Hernia, which is an extension of the 8 common disease patterns listed in our CVPR2017 paper. Note that original radiology reports (associated with these chest x-ray studies) are not meant to be publicly shared for many reasons. The text-mined disease labels are expected to have accuracy >90%.Please find more details and benchmark performance of trained models based on 14 disease labels in our arxiv paper: [1705.02315](https://arxiv.org/abs/1705.02315)_
## Dataset Structure
### Data Instances
A sample from the training set is provided below:
```
{'image_file_path': '/root/.cache/huggingface/datasets/downloads/extracted/95db46f21d556880cf0ecb11d45d5ba0b58fcb113c9a0fff2234eba8f74fe22a/images/00000798_022.png',
'image': <PIL.PngImagePlugin.PngImageFile image mode=L size=1024x1024 at 0x7F2151B144D0>,
'labels': [9, 3]}
```
### Data Fields
The data instances have the following fields:
- `image_file_path` a `str` with the image path
- `image`: A `PIL.Image.Image` object containing the image. Note that when accessing the image column: `dataset[0]["image"]` the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the `"image"` column, *i.e.* `dataset[0]["image"]` should **always** be preferred over `dataset["image"][0]`.
- `labels`: an `int` classification label.
<details>
<summary>Class Label Mappings</summary>
```json
{
"No Finding": 0,
"Atelectasis": 1,
"Cardiomegaly": 2,
"Effusion": 3,
"Infiltration": 4,
"Mass": 5,
"Nodule": 6,
"Pneumonia": 7,
"Pneumothorax": 8,
"Consolidation": 9,
"Edema": 10,
"Emphysema": 11,
"Fibrosis": 12,
"Pleural_Thickening": 13,
"Hernia": 14
}
```
</details>
**Label distribution on the dataset:**
| labels | obs | freq |
|:-------------------|------:|-----------:|
| No Finding | 60361 | 0.426468 |
| Infiltration | 19894 | 0.140557 |
| Effusion | 13317 | 0.0940885 |
| Atelectasis | 11559 | 0.0816677 |
| Nodule | 6331 | 0.0447304 |
| Mass | 5782 | 0.0408515 |
| Pneumothorax | 5302 | 0.0374602 |
| Consolidation | 4667 | 0.0329737 |
| Pleural_Thickening | 3385 | 0.023916 |
| Cardiomegaly | 2776 | 0.0196132 |
| Emphysema | 2516 | 0.0177763 |
| Edema | 2303 | 0.0162714 |
| Fibrosis | 1686 | 0.0119121 |
| Pneumonia | 1431 | 0.0101104 |
| Hernia | 227 | 0.00160382 |
### Data Splits
| |train| test|
|-------------|----:|----:|
|# of examples|86524|25596|
**Label distribution by dataset split:**
| labels | ('Train', 'obs') | ('Train', 'freq') | ('Test', 'obs') | ('Test', 'freq') |
|:-------------------|-------------------:|--------------------:|------------------:|-------------------:|
| No Finding | 50500 | 0.483392 | 9861 | 0.266032 |
| Infiltration | 13782 | 0.131923 | 6112 | 0.164891 |
| Effusion | 8659 | 0.082885 | 4658 | 0.125664 |
| Atelectasis | 8280 | 0.0792572 | 3279 | 0.0884614 |
| Nodule | 4708 | 0.0450656 | 1623 | 0.0437856 |
| Mass | 4034 | 0.038614 | 1748 | 0.0471578 |
| Consolidation | 2852 | 0.0272997 | 1815 | 0.0489654 |
| Pneumothorax | 2637 | 0.0252417 | 2665 | 0.0718968 |
| Pleural_Thickening | 2242 | 0.0214607 | 1143 | 0.0308361 |
| Cardiomegaly | 1707 | 0.0163396 | 1069 | 0.0288397 |
| Emphysema | 1423 | 0.0136211 | 1093 | 0.0294871 |
| Edema | 1378 | 0.0131904 | 925 | 0.0249548 |
| Fibrosis | 1251 | 0.0119747 | 435 | 0.0117355 |
| Pneumonia | 876 | 0.00838518 | 555 | 0.0149729 |
| Hernia | 141 | 0.00134967 | 86 | 0.00232012 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### License and attribution
There are no restrictions on the use of the NIH chest x-ray images. However, the dataset has the following attribution requirements:
- Provide a link to the NIH download site: https://nihcc.app.box.com/v/ChestXray-NIHCC
- Include a citation to the CVPR 2017 paper (see Citation information section)
- Acknowledge that the NIH Clinical Center is the data provider
### Citation Information
```
@inproceedings{Wang_2017,
doi = {10.1109/cvpr.2017.369},
url = {https://doi.org/10.1109%2Fcvpr.2017.369},
year = 2017,
month = {jul},
publisher = {{IEEE}
},
author = {Xiaosong Wang and Yifan Peng and Le Lu and Zhiyong Lu and Mohammadhadi Bagheri and Ronald M. Summers},
title = {{ChestX}-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases},
booktitle = {2017 {IEEE} Conference on Computer Vision and Pattern Recognition ({CVPR})}
}
```
### Contributions
Thanks to [@alcazar90](https://github.com/alcazar90) for adding this dataset.
|
false | # Dataset Card for "oxford-flowers"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards) |
true |
# Dataset Card for Conceptnet5
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**
https://github.com/commonsense/conceptnet5/wiki
- **Repository:**
https://github.com/commonsense/conceptnet5/wiki
- **Paper:**
Robyn Speer, Joshua Chin, and Catherine Havasi. 2017. "ConceptNet 5.5: An Open Multilingual Graph of General Knowledge." In proceedings of AAAI 31.o
### Dataset Summary
ConceptNet is a multilingual knowledge base, representing words and
phrases that people use and the common-sense relationships between
them. The knowledge in ConceptNet is collected from a variety of
resources, including crowd-sourced resources (such as Wiktionary and
Open Mind Common Sense), games with a purpose (such as Verbosity and
nadya.jp), and expert-created resources (such as WordNet and JMDict).
You can browse what ConceptNet knows at http://conceptnet.io.
This dataset is designed to provide training data
for common sense relationships pulls together from various sources.
The dataset is multi-lingual. See langauge codes and language info
here: https://github.com/commonsense/conceptnet5/wiki/Languages
This dataset provides an interface for the conceptnet5 csv file, and
some (but not all) of the raw text data used to build conceptnet5:
omcsnet_sentences_free.txt, and omcsnet_sentences_more.txt.
One use of this dataset would be to learn to extract the conceptnet
relationship from the omcsnet sentences.
Conceptnet5 has 34,074,917 relationships. Of those relationships,
there are 2,176,099 surface text sentences related to those 2M
entries.
omcsnet_sentences_free has 898,161 lines. omcsnet_sentences_more has
2,001,736 lines.
Original downloads are available here
https://github.com/commonsense/conceptnet5/wiki/Downloads. For more
information, see: https://github.com/commonsense/conceptnet5/wiki
The omcsnet data comes with the following warning from the authors of
the above site:
Remember: this data comes from various forms of
crowdsourcing. Sentences in these files are not necessarily true,
useful, or appropriate.
### Languages
en, fr, it, de, es, ru, pt, ja, nl, zh and others
## Dataset Structure
### Data Instances
There are three configurations for the dataset: conceptnet5, omcs_sentences_free, omcs_sentences_more.
Conceptnet5 defines:
``
{
'sentence': ...,
'full_rel': ...,
'rel': ...,
'arg1': ...,
'arg2': ...,
'lang': ...,
'extra_info': ...
'weight': ...
}
``
The omcs text defines:
``
{
'sentence': ...,
'raw_data': ...
'weight': ...
}
``
### Data Fields
For conceptnet5 configurations:
* full_rel: the full relationship. e.g., /a/[/r/Antonym/,/c/en/able/,/c/en/cane/]
* rel: the binary relationship. e.g., /r/Antonym
* arg1: the first argument to the binary relationship. e.g., /c/en/able
* arg2: the second argument to the binary relationship. e.g., /c/en/cane
* lang: the language code. e.g., en, fr, etc. If the arg1 and arg2 are two different languages, then the form os lang1/lang2.
* extra_info: a string that includes json data that has the dataset name, license type (mostly cc-4.0), contributor, etc. e.g., : {"dataset": "/d/verbosity", "license": "cc:by/4.0", "sources": [{"contributor": "/s/resource/verbosity"}], "surfaceEnd": "cane", "surfaceStart": "able", "surfaceText": "[[able]] is the opposite of [[cane]]", "weight": 0.299}
* sentence: the sentence from which the relationship was extracted, if one exists, with brackets around the arg1 and arg2. e.g., [[able]] is the opposite of [[cane]]
* weight: the weight assigned by the curators or automatically to the relationship, between 1.0-0.0, higher being more certain.
For the omcs text configurations:
* sentence: the raw sentence
* raw_data: the raw tab seperated data of the form, id, text, curator_id, created_on, lanugage_id, activity_id, and score. Most of this information was tied to older systems for entering the data os was not partsed into fields for the dataset. e.g., 1237278 someone can be at catch 10805 2006-11-14 17:56:49.70872-05 en 27 1
* lang: the language code
### Data Splits
There are no splits.
## Dataset Creation
### Curation Rationale
This dataset was gathered and created over many years for research in common sense reasoning.
### Source Data
#### Initial Data Collection and Normalization
Started as the Open Mind Common Sense project at MIT Media Lab in 1999. See https://en.wikipedia.org/wiki/Open_Mind_Common_Sense
#### Who are the source language producers?
Crowd Sourced
### Annotations
#### Annotation process
Crowd Source template text, games, etc.
#### Who are the annotators?
Crowd sourced.
### Personal and Sensitive Information
Unkown, but likely there are names of famous individuals.
## Considerations for Using the Data
### Social Impact of Dataset
The goal for the work is to help machines understand common sense.
### Discussion of Biases
See the website and paper for efforts to minimize data bias, but
please note that omcs_sentences_free, omcs_sentences_more are raw data
entered by users and may very well have biased data.
### Other Known Limitations
While the relationship dataset is large, the amount of actual sentences is limited.
## Additional Information
### Dataset Curators
The authors of https://github.com/commonsense/conceptnet5/wiki and Luminoso.
### Licensing Information
This work includes data from ConceptNet 5, which was compiled by the
Commonsense Computing Initiative. ConceptNet 5 is freely available under
the Creative Commons Attribution-ShareAlike license (CC BY SA 3.0) from
http://conceptnet.io.
The included data was created by contributors to Commonsense Computing
projects, contributors to Wikimedia projects, DBPedia, OpenCyc, Games
with a Purpose, Princeton University's WordNet, Francis Bond's Open
Multilingual WordNet, and Jim Breen's JMDict.
Credits and acknowledgements
ConceptNet has been developed by:
The MIT Media Lab, through various groups at different times:
Commonsense Computing
Software Agents
Digital Intuition
The Commonsense Computing Initiative, a worldwide collaboration with contributions from:
National Taiwan University
Universidade Federal de São Carlos
Hokkaido University
Tilburg University
Nihon Unisys Labs
Dentsu Inc.
Kyoto University
Yahoo Research Japan
Luminoso Technologies, Inc.
Significant amounts of data were imported from:
WordNet, a project of Princeton University
Open Multilingual WordNet, compiled by Francis Bond and Kyonghee Paik
Wikipedia and Wiktionary, collaborative projects of the Wikimedia Foundation
Luis von Ahn's "Games with a Purpose"
JMDict, compiled by Jim Breen
CC-CEDict, by MDBG
The Unicode CLDR
DBPedia
Here is a short, incomplete list of people who have made significant contributions to the development of ConceptNet as a data resource, roughly in order of appearance:
Push Singh
Catherine Havasi
Hugo Liu
Hyemin Chung
Robyn Speer
Ken Arnold
Yen-Ling Kuo
Joshua Chin
Joanna Lowry-Duda
Robert Beaudoin
Naoki Otani
Vanya Cohen
Licenses for included resources
Commonsense Computing
The Commonsense Computing project originated at the MIT Media Lab and expanded worldwide. Tens of thousands of contributors have taken some time to teach facts to computers. Their pseudonyms can be found in the "sources" list found in ConceptNet's raw data and in its API.
Games with a Purpose
Data collected from Verbosity, one of the CMU "Games with a Purpose", is used and released under ConceptNet's license, by permission from Luis von Ahn and Harshit Surana.
Verbosity players are anonymous, so in the "sources" list, data from Verbosity is simply credited to the pseudonym "verbosity".
Wikimedia projects
ConceptNet uses data directly from Wiktionary, the free dictionary. It also uses data from Wikipedia, the free encyclopedia via DBPedia.
Wiktionary and Wikipedia are collaborative projects, authored by their respective online communities. They are currently released under the Creative Commons Attribution-ShareAlike license.
Wikimedia encourages giving attribution by providing links to the hosted pages that the data came from, and DBPedia asks for the same thing in turn. In addition to crediting the assertions that came from Wiktionary and DBPedia, we also provide "ExternalURL" edges pointing to the page that they came from. For example, the term /c/de/sprache has an ExternalURL link pointing to http://en.wiktionary.org/wiki/Sprache. Its list of individual contributors can be seen by following its "History" link.
The URLs of links to DBPedia are the same as the resource names that DBPedia uses, encouraging interoperability with their linked data.
WordNet
WordNet is available under an unencumbered license: see http://wordnet.princeton.edu/wordnet/license/. Its text is reproduced below:
WordNet Release 3.0
This software and database is being provided to you, the LICENSEE, by Princeton University under the following license. By obtaining, using and/or copying this software and database, you agree that you have read, understood, and will comply with these terms and conditions.:
Permission to use, copy, modify and distribute this software and database and its documentation for any purpose and without fee or royalty is hereby granted, provided that you agree to comply with the following copyright notice and statements, including the disclaimer, and that the same appear on ALL copies of the software, database and documentation, including modifications that you make for internal use or for distribution.
WordNet 3.0 Copyright 2006 by Princeton University. All rights reserved.
THIS SOFTWARE AND DATABASE IS PROVIDED "AS IS" AND PRINCETON UNIVERSITY MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR IMPLIED. BY WAY OF EXAMPLE, BUT NOT LIMITATION, PRINCETON UNIVERSITY MAKES NO REPRESENTATIONS OR WARRANTIES OF MERCHANT- ABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF THE LICENSED SOFTWARE, DATABASE OR DOCUMENTATION WILL NOT INFRINGE ANY THIRD PARTY PATENTS, COPYRIGHTS, TRADEMARKS OR OTHER RIGHTS.
The name of Princeton University or Princeton may not be used in advertising or publicity pertaining to distribution of the software and/or database. Title to copyright in this software, database and any associated documentation shall at all times remain with Princeton University and LICENSEE agrees to preserve same.
Open Multilingual WordNet
Open Multilingual WordNet was compiled by Francis Bond, Kyonghee Paik, and Ryan Foster, from data provided by many multilingual WordNet projects. Here is the complete list of references to the projects that created the data.
### Citation Information
Robyn Speer, Joshua Chin, and Catherine Havasi. 2017. "ConceptNet 5.5: An Open Multilingual Graph of General Knowledge." In proceedings of AAAI 31.
### Contributions
Thanks to [@ontocord](https://github.com/ontocord) for adding this dataset. |
false | # LongForm
The LongForm dataset is created by leveraging English corpus
examples with augmented instructions. We select a
diverse set of human-written
documents from existing corpora such as C4 and
Wikipedia and generate instructions for the given
documents via LLMs. Then, we extend these examples with structured corpora examples such as Stack Exchange and WikiHow and task examples such as question answering, email writing, grammar error correction, story/poem generation, and text summarization.
## Distribution
The distribution of the LongForm dataset in terms of the source of examples is below. It contains examples generated from raw text corpora via LLMs, structured corpus examples, as well as various NLP task examples such as email writing, grammar error correction, story/poem generation, and text summarization.
| **Type** | **Source** | **Number of Examples** |
|------------------------|----------------|------------------------|
| **Corpora** | C4 | 10,000 |
| | Wikipedia | 5,000 |
| **Structured Corpora** | Stack Exchange | 4,380 |
| | WikiHow | 2,500 |
| **Tasks** | NIv2 | 3,684 |
| | Big Bench | 600 |
| | BEA-GEC | 1,203 |
| | Enron | 372 |
| **Total** | | 27,739 |
| | | |
| **Train** | | 23,652 |
| **Validation** | | 2,042 |
| **Test** | | 2,045 |
## Models
| | **All** | **Recipe Generation** | **ELI5** | **Writing Prompts** |
|-----------------------|---------|-----------------------------------|----------|---------------------|
| **T0++** | 10.9 | 18.7 | 3.8 | 10.2 |
| **Tk-Instruct** | 6.3 | 12.9* | 3.6 | 2.4 |
| **Flan-T5** | 10.6 | 20.9* | 3.5 | 7.4 |
| **Alpaca-LLaMA-7B** | 14.6 | 19.5 | 12.5 | 11.8 |
| **OPT-30B** | 11.1 | 18.6 | 12.2 | 2.6 |
| **[LongForm-T5-XL](https://huggingface.co/akoksal/LongForm-T5-XL)** | 16.3 | 20.2 | 18.3 | 10.6 |
| **[LongForm-OPT-2.7B](https://huggingface.co/akoksal/LongForm-OPT-2.7B)** | 17.8 | 15.5 | 17.9 | **19.9** |
| **[LongForm-OPT-6.7B](https://huggingface.co/akoksal/LongForm-OPT-6.7B)** | 17.7 | 16.9 | 17.2 | 19.0 |
| [**LongForm-LLaMA-7B**](https://huggingface.co/akoksal/LongForm-LLaMA-7B-diff)‡ | **19.7** | **21.7** | **18.6** | 18.9 |
‡: We can just release the difference between LongForm-LLaMA-7B and pretrained LLaMA-7B publicly due to restrictions of LLaMA models.
## Examples
| **Source** | **Instruction** | **Output**|
|------------------------|----------------|------------------------|
| C4 | What is the likelihood that Donald Trump's HHS secretary nominee, Alex Azar, will push policies that benefit pharmaceutical companies' pocketbooks? | His record suggests he’s unlikely to push pricing policies that would hit pharmaceutical companies’ pocketbooks.<br />President Donald Trump tweeted Monday that his nominee for HHS secretary, Alex Azar will “be a star for … lower drug prices!” But the record of the former top executive for Eli Lilly, which tripled the price of a top-selling insulin drug while he led its U.S. operation, suggests a different story.<br />Lilly is one of three drug companies targeted by a class-action lawsuit that accuses the company[...]<br />“This nominee has impeccable big pharma credentials but non-existent credentials when it comes to concerns on drug prices,” said Peter Welch, co-chair of the House Democrats Drug Pricing Task Force.
|Wikipedia | "Where is San Saba located in Rome, Italy?" Respond in 2 sentences. |San Saba is an ancient basilica church in Rome, Italy. It lies on the so-called "Piccolo Aventino", which is an area close to the ancient Aurelian Walls next to the Aventine Hill and Caelian Hill.|
## Dataset Description
- **Repository:** https://github.com/akoksal/LongForm
- **Paper:** https://arxiv.org/abs/2304.08460
- **Version:** v1.0 - April 18, 2023
- **Contact:** [Abdullatif Köksal](https://twitter.com/akoksal_)
## License
The LongForm project is subject to a MIT License with custom limitations for restrictions imposed by OpenAI (for the instruction generation part), as well as the license of language models (OPT, LLaMA, and T5).
## Citation
```
@misc{koksal2023longform,
title={LongForm: Optimizing Instruction Tuning for Long Text Generation with Corpus Extraction},
author={Abdullatif Köksal and Timo Schick and Anna Korhonen and Hinrich Schütze},
year={2023},
eprint={2304.08460},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
false |
# Dataset Card for [Persian NER]
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Github](https://github.com/HaniehP/PersianNER)
- **Repository:** [Github](https://github.com/HaniehP/PersianNER)
- **Paper:** [Aclweb](https://www.aclweb.org/anthology/C16-1319)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
The dataset includes 7,682 Persian sentences, split into 250,015 tokens and their NER labels. It is available in 3 folds to be used in turn as training and test sets. The NER tags are in IOB format.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
### Data Fields
- `id`: id of the sample
- `tokens`: the tokens of the example text
- `ner_tags`: the NER tags of each token
The NER tags correspond to this list:
```
"O", "I-event", "I-fac", "I-loc", "I-org", "I-pers", "I-pro", "B-event", "B-fac", "B-loc", "B-org", "B-pers", "B-pro"
```
### Data Splits
Training and test splits
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
Hanieh Poostchi, Ehsan Zare Borzeshi, Mohammad Abdous, Massimo Piccardi
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
Hanieh Poostchi, Ehsan Zare Borzeshi, Mohammad Abdous, Massimo Piccardi
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
Dataset is published for academic use only
### Dataset Curators
[More Information Needed]
### Licensing Information
Creative Commons Attribution 4.0 International License.
### Citation Information
@inproceedings{poostchi-etal-2016-personer,
title = "{P}erso{NER}: {P}ersian Named-Entity Recognition",
author = "Poostchi, Hanieh and
Zare Borzeshi, Ehsan and
Abdous, Mohammad and
Piccardi, Massimo",
booktitle = "Proceedings of {COLING} 2016, the 26th International Conference on Computational Linguistics: Technical Papers",
month = dec,
year = "2016",
address = "Osaka, Japan",
publisher = "The COLING 2016 Organizing Committee",
url = "https://www.aclweb.org/anthology/C16-1319",
pages = "3381--3389",
abstract = "Named-Entity Recognition (NER) is still a challenging task for languages with low digital resources. The main difficulties arise from the scarcity of annotated corpora and the consequent problematic training of an effective NER pipeline. To abridge this gap, in this paper we target the Persian language that is spoken by a population of over a hundred million people world-wide. We first present and provide ArmanPerosNERCorpus, the first manually-annotated Persian NER corpus. Then, we introduce PersoNER, an NER pipeline for Persian that leverages a word embedding and a sequential max-margin classifier. The experimental results show that the proposed approach is capable of achieving interesting MUC7 and CoNNL scores while outperforming two alternatives based on a CRF and a recurrent neural network.",
}
### Contributions
Thanks to [@KMFODA](https://github.com/KMFODA) for adding this dataset. |
false |
# Dataset Card for [Dataset Name]
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [PT Gria Inovasi Teknologi](https://grit.id/)
- **Repository:** [Nergrit Corpus](https://github.com/grit-id/nergrit-corpus)
- **Paper:**
- **Leaderboard:**
- **Point of Contact:** [Taufiqur Rohman](mailto:taufiq@grit.id)
### Dataset Summary
Nergrit Corpus is a dataset collection of Indonesian Named Entity Recognition, Statement Extraction,
and Sentiment Analysis developed by [PT Gria Inovasi Teknologi (GRIT)](https://grit.id/).
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
Indonesian
## Dataset Structure
A data point consists of sentences seperated by empty line and tab-seperated tokens and tags.
```
{'id': '0',
'tokens': ['Gubernur', 'Bank', 'Indonesia', 'menggelar', 'konferensi', 'pers'],
'ner_tags': [9, 28, 28, 38, 38, 38],
}
```
### Data Instances
[More Information Needed]
### Data Fields
- `id`: id of the sample
- `tokens`: the tokens of the example text
- `ner_tags`: the NER tags of each token
#### Named Entity Recognition
The ner_tags correspond to this list:
```
"B-CRD", "B-DAT", "B-EVT", "B-FAC", "B-GPE", "B-LAN", "B-LAW", "B-LOC", "B-MON", "B-NOR",
"B-ORD", "B-ORG", "B-PER", "B-PRC", "B-PRD", "B-QTY", "B-REG", "B-TIM", "B-WOA",
"I-CRD", "I-DAT", "I-EVT", "I-FAC", "I-GPE", "I-LAN", "I-LAW", "I-LOC", "I-MON", "I-NOR",
"I-ORD", "I-ORG", "I-PER", "I-PRC", "I-PRD", "I-QTY", "I-REG", "I-TIM", "I-WOA", "O",
```
The ner_tags have the same format as in the CoNLL shared task: a B denotes the first item of a phrase and an I any
non-initial word. The dataset contains 19 following entities
```
'CRD': Cardinal
'DAT': Date
'EVT': Event
'FAC': Facility
'GPE': Geopolitical Entity
'LAW': Law Entity (such as Undang-Undang)
'LOC': Location
'MON': Money
'NOR': Political Organization
'ORD': Ordinal
'ORG': Organization
'PER': Person
'PRC': Percent
'PRD': Product
'QTY': Quantity
'REG': Religion
'TIM': Time
'WOA': Work of Art
'LAN': Language
```
#### Sentiment Analysis
The ner_tags correspond to this list:
```
"B-NEG", "B-NET", "B-POS",
"I-NEG", "I-NET", "I-POS",
"O",
```
#### Statement Extraction
The ner_tags correspond to this list:
```
"B-BREL", "B-FREL", "B-STAT", "B-WHO",
"I-BREL", "I-FREL", "I-STAT", "I-WHO",
"O"
```
The ner_tags have the same format as in the CoNLL shared task: a B denotes the first item of a phrase and an I any
non-initial word.
### Data Splits
The dataset is splitted in to train, validation and test sets.
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
The annotators are listed in the
[Nergrit Corpus repository](https://github.com/grit-id/nergrit-corpus)
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@cahya-wirawan](https://github.com/cahya-wirawan) for adding this dataset. |
false |
# Dataset Card for [Dataset Name]
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Interactive Demo:** [Interactive demo](http://protoqa.com)
- **Repository:** [proto_qa repository](https://github.com/iesl/protoqa-data)
- **Paper:** [proto_qa paper](https://arxiv.org/pdf/2005.00771.pdf)
- **Point of Contact:** [Michael Boratko](mailto:mboratko@cs.umass.edu)
[Xiang Lorraine Li](mailto:xiangl@cs.umass.edu)
[Tim O’Gorman](mailto:togorman@cs.umass.edu)
[Rajarshi Das](mailto:rajarshi@cs.umass.edu)
[Dan Le](mailto:dhle@cs.umass.edu)
[Andrew McCallum](mailto:mccallum@cs.umass.edu)
### Dataset Summary
This dataset is for studying computational models trained to reason about prototypical situations. It is anticipated that still would not lead to usage in a downstream task, but as a way of studying the knowledge (and biases) of prototypical situations already contained in pre-trained models. The data it is partially based on (Family Feud).
Using deterministic filtering a sampling from a larger set of all transcriptions was built. Scraped data was acquired through fan transcriptions at [family feud](https://www.familyfeudinfo.com) and [family feud friends](http://familyfeudfriends.arjdesigns.com/); crowdsourced data was acquired with FigureEight (now Appen)
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
The text in the dataset is in English
## Dataset Structure
### Data Instances
**What do the instances that comprise the dataset represent?**<br>
Each represents a survey question from Family Feud game and reported answer clusters
**How many instances are there in total?**<br>
9789 instances
**What data does each instance consist of?**<br>
Each instance is a question, a set of answers, and a count associated with each answer.
### Data Fields
**Data Files**<br>
Each line is a json dictionary, in which:<br>
**question** contains the question (in original and a normalized form)<br>
**answerstrings** contains the original answers provided by survey respondents (when available), along with the counts for each string. Because the FamilyFeud data has only cluster names rather than strings, those cluster names are included with 0 weight.<br>
**answer-clusters** list of clusters, with the count of each cluster and the strings included in that cluster. Each cluster is given a unique ID that can be linked to in the assessment files.
The simplified configuration includes:
- `question`: contains the original question
- `normalized-question`: contains the question in normalized form
- `totalcount`: unique identifier of the comment (can be used to look up the entry in the raw dataset)
- `id`: unique identifier of the commen
- `source`: unique identifier of the commen
- `answerstrings`: unique identifier of the commen
- `answer-clusters | answers-cleaned`: list clusters of:
* `clusterid`: Each cluster is given a unique ID that can be linked to in the assessment files
* `count`: the count of each cluster
* `answers`: the strings included in that cluster
In addition to the above, there is crowdsourced assessments file. The config "proto_qa_cs_assessments" provides mappings from additional human and model answers to clusters, to evaluate different assessment methods.
**Assessment files**<br>
The file **data/dev/crowdsource_dev.assessments.jsonl** contains mappings from additional human and model answers to clusters, to evaluate different assessment methods.
Each line contains:<br>
* `question`: contains the ID of the question
* `assessments`: maps individual strings to one of three options, either the answer cluster id, "invalid" if the answer is judged to be bad, or "valid_new_cluster" if the answer is valid but does not match any existing clusters.
### Data Splits
* proto_qa `Train` : 8781 instances for training or fine-tuning scraped from Family Feud fan sites (see paper). Scraped data has answer clusters with sizes, but only has a single string per cluster (corresponding to the original cluster name
* proto_qa `Validation` : 979 instances sampled from the same Family Feud data, for use in model validation and development.
* proto_qa_cs `Validation` :: 51 questions collected with exhaustive answer collection and manual clustering, matching the details of the eval test set (roughly 100 human answers per question)
**data/dev/crowdsource_dev.assessments.jsonl**: assessment file (format described above) for study of assessment methods.
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
**How was the data associated with each instance acquired?**<br>
Scraped data was acquired through fan transcriptions at https://www.familyfeudinfo.com and http://familyfeudfriends.arjdesigns.com/ ; crowdsourced data was acquired with FigureEight (now Appen)
**If the dataset is a sample from a larger set, what was the sampling strategy?**<br>
Deterministic filtering was used (noted elsewhere), but no probabilistic sampling was used.
**Who was involved in the data collection process (e.g., students,crowdworkers , contractors) and how were they compensated?**<br>
Crowdworkers were used in the evalaution dataset. Time per task was calculated and per-task cost was set to attempt to provide a living wage
**Over what timeframe was the data collected?**<br>
Crowdsource answers were collected between Fall of 2018 and Spring of 2019. Scraped data covers question-answer pairs collected since the origin of the show in 1976
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
**Was any preprocessing/cleaning/labeling of the data done?**<br>
Obvious typos in the crowdsourced answer set were corrected
#### Who are the annotators?
The original question-answer pairs were generated by surveys of US English-speakers in a period from 1976 to present day. Crowd-sourced evaluation was constrained geographically to US English speakers but not otherwise constrained. Additional demographic data was not collected.
### Personal and Sensitive Information
**Does the dataset contain data that might be considered sensitive in any way?**<br>
As the questions address prototypical/stereotypical activities, models trained on more offensive material (such as large language models) may provide offensive answers to such questions. While we had found a few questions which we worried would actually encourage models to provide offensive answers, we cannot guarantee that the data is clean of such questions. Even a perfectly innocent version of this dataset would be encouraging models to express generalizations about situations, and therefore may provoke offensive material that is oontained in language models
**Does the dataset contain data that might be considered confidential?**<br>
The data does not concern individuals and thus does not contain any information to identify persons. Crowdsourced answers do not provide any user identifiers.
## Considerations for Using the Data
### Social Impact of Dataset
**Does the dataset contain data that, if viewed directly, might be offensive, insulting, threatening, or might otherwise cause anxiety?**<br>
Not egregiously so (questions are all designed to be shown on television or replications thereof),
### Discussion of Biases
**Is there anything about the composition of the dataset or the way it was collected and preprocessed/cleaned/labeled that might impact future uses?**
<br>All original questions were written with US television audiences in mind, and therefore characterize prototypical situations with a specific lens. Any usages which deploy this to actually model prototypical situations globally will carry that bias.
**Are there tasks for which the dataset should not be used?**
<br>We caution regarding free-form use of this dataset for interactive "commonsense question answering" purposes without more study of the biases and stereotypes learned by such models.
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
The listed authors are maintaining/supporting the dataset. They pledge to help support issues, but cannot guarantee long-term support
### Licensing Information
The Proto_qa dataset is licensed under the [Creative Commons Attribution 4.0 International](https://github.com/iesl/protoqa-data/blob/master/LICENSE)
### Citation Information
```
@InProceedings{
huggingface:dataset,
title = {ProtoQA: A Question Answering Dataset for Prototypical Common-Sense Reasoning},
authors = {Michael Boratko, Xiang Lorraine Li, Tim O’Gorman, Rajarshi Das, Dan Le, Andrew McCallum},
year = {2020},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {https://github.com/iesl/protoqa-data},
}
```
### Contributions
Thanks to [@bpatidar](https://github.com/bpatidar) for adding this dataset. |
false |
# Dataset Card for BEIR Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/UKPLab/beir
- **Repository:** https://github.com/UKPLab/beir
- **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
- **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
- **Point of Contact:** nandan.thakur@uwaterloo.ca
### Dataset Summary
BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
- Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
- Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
- Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
- News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
- Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
- Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
- Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
- Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
- Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
All these datasets have been preprocessed and can be used for your experiments.
```python
```
### Supported Tasks and Leaderboards
The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
### Languages
All tasks are in English (`en`).
## Dataset Structure
All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
- `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
- `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
- `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
### Data Instances
A high level example of any beir dataset:
```python
corpus = {
"doc1" : {
"title": "Albert Einstein",
"text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
its influence on the philosophy of science. He is best known to the general public for his mass–energy \
equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
of the photoelectric effect', a pivotal step in the development of quantum theory."
},
"doc2" : {
"title": "", # Keep title an empty string if not present
"text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
},
}
queries = {
"q1" : "Who developed the mass-energy equivalence formula?",
"q2" : "Which beer is brewed with a large proportion of wheat?"
}
qrels = {
"q1" : {"doc1": 1},
"q2" : {"doc2": 1},
}
```
### Data Fields
Examples from all configurations have the following features:
### Corpus
- `corpus`: a `dict` feature representing the document title and passage text, made up of:
- `_id`: a `string` feature representing the unique document id
- `title`: a `string` feature, denoting the title of the document.
- `text`: a `string` feature, denoting the text of the document.
### Queries
- `queries`: a `dict` feature representing the query, made up of:
- `_id`: a `string` feature representing the unique query id
- `text`: a `string` feature, denoting the text of the query.
### Qrels
- `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
- `_id`: a `string` feature representing the query id
- `_id`: a `string` feature, denoting the document id.
- `score`: a `int32` feature, denoting the relevance judgement between query and document.
### Data Splits
| Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
| -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
| MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
| TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
| NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
| BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
| NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
| HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
| FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
| Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
| TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
| ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
| Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
| CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
| Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
| DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
| SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
| FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
| Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
| SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
| Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
Cite as:
```
@inproceedings{
thakur2021beir,
title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
year={2021},
url={https://openreview.net/forum?id=wCu6T5xFjeJ}
}
```
### Contributions
Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset. |
true | # Dataset Card for FigLang2022SharedTask
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://figlang2022sharedtask.github.io/
- **Repository:**
- **Paper:** TBA
- **Point of Contact:** tuhin.chakr@cs.columbia.edu
### Dataset Summary
Model in the loop approach for fig lang generation and explainability
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
TBA
### Contributions
Thanks to [@github-username](https://github.com/<github-username>) for adding this dataset. |
false |
# Dataset Card for "tner/bc5cdr"
## Dataset Description
- **Repository:** [T-NER](https://github.com/asahi417/tner)
- **Paper:** [https://academic.oup.com/database/article/doi/10.1093/database/baw032/2630271?login=true](https://academic.oup.com/database/article/doi/10.1093/database/baw032/2630271?login=true)
- **Dataset:** BioCreative V CDR
- **Domain:** Biomedical
- **Number of Entity:** 2
### Dataset Summary
BioCreative V CDR NER dataset formatted in a part of [TNER](https://github.com/asahi417/tner) project.
The original dataset consists of long documents which cannot be fed on LM because of the length, so we split them into sentences to reduce their size.
- Entity Types: `Chemical`, `Disease`
## Dataset Structure
### Data Instances
An example of `train` looks as follows.
```
{
'tags': [2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0],
'tokens': ['Fasciculations', 'in', 'six', 'areas', 'of', 'the', 'body', 'were', 'scored', 'from', '0', 'to', '3', 'and', 'summated', 'as', 'a', 'total', 'fasciculation', 'score', '.']
}
```
### Label ID
The label2id dictionary can be found at [here](https://huggingface.co/datasets/tner/bc5cdr/raw/main/dataset/label.json).
```python
{
"O": 0,
"B-Chemical": 1,
"B-Disease": 2,
"I-Disease": 3,
"I-Chemical": 4
}
```
### Data Splits
| name |train|validation|test|
|---------|----:|---------:|---:|
|bc5cdr|5228| 5330|5865|
### Citation Information
```
@article{wei2016assessing,
title={Assessing the state of the art in biomedical relation extraction: overview of the BioCreative V chemical-disease relation (CDR) task},
author={Wei, Chih-Hsuan and Peng, Yifan and Leaman, Robert and Davis, Allan Peter and Mattingly, Carolyn J and Li, Jiao and Wiegers, Thomas C and Lu, Zhiyong},
journal={Database},
volume={2016},
year={2016},
publisher={Oxford Academic}
}
``` |
false |
# Dataset Card for MSLR2022
## Table of Contents
- [Dataset Card for MSLR2022](#dataset-card-for-mslr2022)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Initial Data Collection and Normalization](#initial-data-collection-and-normalization)
- [Who are the source language producers?](#who-are-the-source-language-producers)
- [Annotations](#annotations)
- [Annotation process](#annotation-process)
- [Who are the annotators?](#who-are-the-annotators)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** https://github.com/allenai/mslr-shared-task
- **Repository:** https://github.com/allenai/mslr-shared-task
- **Paper:** https://aclanthology.org/2021.emnlp-main.594
- **Leaderboard:** https://github.com/allenai/mslr-shared-task#leaderboard
- **Point of Contact:** https://github.com/allenai/mslr-shared-task#contact-us
### Dataset Summary
The Multidocument Summarization for Literature Review (MSLR) Shared Task aims to study how medical evidence from different clinical studies are summarized in literature reviews. Reviews provide the highest quality of evidence for clinical care, but are expensive to produce manually. (Semi-)automation via NLP may facilitate faster evidence synthesis without sacrificing rigor. The MSLR shared task uses two datasets to assess the current state of multidocument summarization for this task, and to encourage the development of modeling contributions, scaffolding tasks, methods for model interpretability, and improved automated evaluation methods in this domain.
### Supported Tasks and Leaderboards
This dataset is used for the MSLR2022 Shared Task. For information on the shared task leaderboard, please refer [here](https://github.com/allenai/mslr-shared-task#leaderboard).
### Languages
English
## Dataset Structure
More information on dataset structure [here](https://github.com/allenai/mslr-shared-task#data-structure).
### Data Instances
__MS^2__
```json
{
"review_id": "30760312",
"pmid": [
"22776744",
"25271670",
"3493740",
"1863023",
"16291984",
"23984728",
"23996433",
"18466198",
"12151469",
"27400308",
"16053970",
"22922316",
"11897647",
"11597664",
"4230647"
],
"title": [
"Improved Cell Survival and Paracrine Capacity of Human Embryonic Stem Cell-Derived Mesenchymal Stem Cells Promote Therapeutic Potential for Pulmonary Arterial Hypertension",
"Adipose-derived stem cells attenuate pulmonary arterial hypertension and ameliorate pulmonary arterial remodeling in monocrotaline-induced pulmonary hypertensive rats",
"Effect of bone marrow mesenchymal stem cells on experimental pulmonary arterial hypertension",
"Survival in patients with primary pulmonary hypertension. Results from a national prospective registry.",
"Sildenafil citrate therapy for pulmonary arterial hypertension.",
"Macitentan and morbidity and mortality in pulmonary arterial hypertension.",
"Long-term research of stem cells in monocrotaline-induced pulmonary arterial hypertension",
"Safety and efficacy of autologous endothelial progenitor cells transplantation in children with idiopathic pulmonary arterial hypertension: open-label pilot study.",
"Inhaled iloprost for severe pulmonary hypertension.",
"Sildenafil reduces pulmonary vascular resistance in single ventricular physiology.",
"Ambrisentan therapy for pulmonary arterial hypertension.",
"Mesenchymal stem cell prevention of vascular remodeling in high flow-induced pulmonary hypertension through a paracrine mechanism.",
"Continuous subcutaneous infusion of treprostinil, a prostacyclin analogue, in patients with pulmonary arterial hypertension: a double-blind, randomized, placebo-controlled trial.",
"Effects of the dual endothelin-receptor antagonist bosentan in patients with pulmonary hypertension: a randomised placebocontrolled study",
"SYRCLE\\u2019s risk of bias tool for animal studies"
],
"abstract": [
"Although transplantation of adult bone marrow mesenchymal stem cells ( BM-MSCs ) holds promise in the treatment for pulmonary arterial hypertension ( PAH ) , the poor survival and differentiation potential of adult BM-MSCs have limited their therapeutic efficiency . Here , we compared the therapeutic efficacy of human embryonic stem cell-derived MSCs ( hESC-MSCs ) with adult BM-MSCs for the treatment of PAH in an animal model . One week following monocrotaline (MCT)-induced PAH , mice were r and omly assigned to receive phosphate-buffered saline ( MCT group ) ; 3.0 \\u00d7 106 human BM-derived MSCs ( BM-MSCs group ) or 3.0 \\u00d7 106 hESC-derived MSCs ( hESC-MSCs group ) via tail vein injection . At 3 weeks posttransplantation , the right ventricular systolic pressure ( RVSP ) , degree of RV hypertrophy , and medial wall thickening of pulmonary arteries were lower= , and pulmonary capillary density was higher in the hESC-MSC group as compared with BM-MSC and MCT groups ( all p < 0.05 ) . At 1 week posttransplantation , the number of engrafted MSCs in the lungs was found significantly higher in the hESC-MSC group than in the BM-MSC group ( all p < 0.01 ) . At 3 weeks posttransplantation , implanted BM-MSCs were undetectable whereas hESC-MSCs were not only engrafted in injured pulmonary arteries but had also undergone endothelial differentiation . In addition , protein profiling of hESC-MSC- and BM-MSC-conditioned medium revealed a differential paracrine capacity . Classification of these factors into bioprocesses revealed that secreted factors from hESC-MSCs were preferentially involved in early embryonic development and tissue differentiation , especially blood vessel morphogenesis . We concluded that improved cell survival and paracrine capacity of hESC-MSCs provide better therapeutic efficacy than BM-MSCs in the treatment for PAH",
"Abstract We investigated the effect of adipose-derived stem cells ( ADSCs ) transplantation effects on structural remodeling and pulmonary artery pressure in monocrotaline (MCT)-induced pulmonary hypertensive rats . In the first experiment , 32 male Sprague-Dawley ( SD ) rats were r and omly divided into four groups ( n = 8/group ) : 3 ADSCs treated groups and normal control ( Ctrl ) . ADSCs were administered through the left jugular vein at 105 , 106 and 107 cells , respectively , and a cell density of 106cells/ml was shown to be optimal . The GFP-tagged ADSCs were identified in the lungs and differentiated into endothelial-like cells . In the second experiment , 96 male SD rats were r and omly divided into three groups ( n = 32/group ) : Ctrl , MCT-induced pulmonary arterial hypertension ( PAH ) , and PAH treated with ADSCs ( ADSCs ) . Two weeks post-MCT administration , the ADSCs group received 1 \\u00d7 106 ADSCs via the external jugular vein . Compared to PAH rats , mean pulmonary arterial pressure was decreased in rats at 1 , 2 , and 3 weeks after ADSCs-treatment ( 18.63 \\u00b1 2.15 mmHg versus 24.53 \\u00b1 2.90 mmHg ; 23.07 \\u00b1 2.84 mmHg versus 33.18 \\u00b1 2.30 mmHg ; 22.98 \\u00b1 2.34 mmHg versus 36.38 \\u00b1 3.28 mmHg , p < 0.05 ) . Meanwhile , the right heart hypertrophy index ( 36.2 1 \\u00b1 4.27 % versus 41.01 \\u00b1 1.29 % ; 39.47 \\u00b1 4.02 % versus 48.75 \\u00b1 2 .13 % ; 41.02 \\u00b1 0.9 % versus 50.52 \\u00b1 1.49 % , p < 0.05 , respectively ) , ratio of wall/lumen thickness , as well as the wall/lumen area were significantly reduced in PAH rats at these time points following ADSCs-treatment , as compared with untreated PAH rats . In summary , ADSCs may colonize the pulmonary arteries , attenuate pulmonary arterial hypertension and ameliorate pulmonary arterial remodeling",
"The aim of the present study was to investigate the effect of bone marrow mesenchymal stem cell ( BMSC ) transp1antation on lung and heart damage in a rat model of monocrotaline (MCT)-induced pulmonary arterial hypertension ( PAH ) . The animals were r and omly divided into 3 groups : control , PAH and BMSC implantation groups . Structural changes in the pulmonary vascular wall , such as the pulmonary artery lumen area ( VA ) and vascular area ( TAA ) were measured by hematoxylin and eosin ( H&E ) staining , and the hemodynamics were detected by echocardiography . Two weeks post-operation , our results demonstrated that sublingual vein injection of BMSCs significantly attenuated the pulmonary vascular structural and hemodynamic changes caused by pulmonary arterial hypertension . The mechanism may be executed via paracrine effects",
"OBJECTIVE To characterize mortality in persons diagnosed with primary pulmonary hypertension and to investigate factors associated with survival . DESIGN Registry with prospect i ve follow-up . SETTING Thirty-two clinical centers in the United States participating in the Patient Registry for the Characterization of Primary Pulmonary Hypertension supported by the National Heart , Lung , and Blood Institute . PATIENTS Patients ( 194 ) diagnosed at clinical centers between 1 July 1981 and 31 December 1985 and followed through 8 August 1988 . MEASUREMENTS At diagnosis , measurements of hemodynamic variables , pulmonary function , and gas exchange variables were taken in addition to information on demographic variables , medical history , and life-style . Patients were followed for survival at 6-month intervals . MAIN RESULTS The estimated median survival of these patients was 2.8 years ( 95 % Cl , 1.9 to 3.7 years ) . Estimated single-year survival rates were as follows : at 1 year , 68 % ( Cl , 61 % to 75 % ) ; at 3 years , 48 % ( Cl , 41 % to 55 % ) ; and at 5 years , 34 % ( Cl , 24 % to 44 % ) . Variables associated with poor survival included a New York Heart Association ( NYHA ) functional class of III or IV , presence of Raynaud phenomenon , elevated mean right atrial pressure , elevated mean pulmonary artery pressure , decreased cardiac index , and decreased diffusing capacity for carbon monoxide ( DLCO ) . Drug therapy at entry or discharge was not associated with survival duration . CONCLUSIONS Mortality was most closely associated with right ventricular hemodynamic function and can be characterized by means of an equation using three variables : mean pulmonary artery pressure , mean right atrial pressure , and cardiac index . Such an equation , once vali date d prospect ively , could be used as an adjunct in planning treatment strategies and allocating medical re sources",
"BACKGROUND Sildenafil inhibits phosphodiesterase type 5 , an enzyme that metabolizes cyclic guanosine monophosphate , thereby enhancing the cyclic guanosine monophosphate-mediated relaxation and growth inhibition of vascular smooth-muscle cells , including those in the lung . METHODS In this double-blind , placebo-controlled study , we r and omly assigned 278 patients with symptomatic pulmonary arterial hypertension ( either idiopathic or associated with connective-tissue disease or with repaired congenital systemic-to-pulmonary shunts ) to placebo or sildenafil ( 20 , 40 , or 80 mg ) orally three times daily for 12 weeks . The primary end point was the change from baseline to week 12 in the distance walked in six minutes . The change in mean pulmonary-artery pressure and World Health Organization ( WHO ) functional class and the incidence of clinical worsening were also assessed , but the study was not powered to assess mortality . Patients completing the 12-week r and omized study could enter a long-term extension study . RESULTS The distance walked in six minutes increased from baseline in all sildenafil groups ; the mean placebo-corrected treatment effects were 45 m ( + 13.0 percent ) , 46 m ( + 13.3 percent ) , and 50 m ( + 14.7 percent ) for 20 , 40 , and 80 mg of sildenafil , respectively ( P<0.001 for all comparisons ) . All sildenafil doses reduced the mean pulmonary-artery pressure ( P=0.04 , P=0.01 , and P<0.001 , respectively ) , improved the WHO functional class ( P=0.003 , P<0.001 , and P<0.001 , respectively ) , and were associated with side effects such as flushing , dyspepsia , and diarrhea . The incidence of clinical worsening did not differ significantly between the patients treated with sildenafil and those treated with placebo . Among the 222 patients completing one year of treatment with sildenafil monotherapy , the improvement from baseline at one year in the distance walked in six minutes was 51 m. CONCLUSIONS Sildenafil improves exercise capacity , WHO functional class , and hemodynamics in patients with symptomatic pulmonary arterial hypertension",
"BACKGROUND Current therapies for pulmonary arterial hypertension have been adopted on the basis of short-term trials with exercise capacity as the primary end point . We assessed the efficacy of macitentan , a new dual endothelin-receptor antagonist , using a primary end point of morbidity and mortality in a long-term trial . METHODS We r and omly assigned patients with symptomatic pulmonary arterial hypertension to receive placebo once daily , macitentan at a once-daily dose of 3 mg , or macitentan at a once-daily dose of 10 mg . Stable use of oral or inhaled therapy for pulmonary arterial hypertension , other than endothelin-receptor antagonists , was allowed at study entry . The primary end point was the time from the initiation of treatment to the first occurrence of a composite end point of death , atrial septostomy , lung transplantation , initiation of treatment with intravenous or subcutaneous prostanoids , or worsening of pulmonary arterial hypertension . RESULTS A total of 250 patients were r and omly assigned to placebo , 250 to the 3-mg macitentan dose , and 242 to the 10-mg macitentan dose . The primary end point occurred in 46.4 % , 38.0 % , and 31.4 % of the patients in these groups , respectively . The hazard ratio for the 3-mg macitentan dose as compared with placebo was 0.70 ( 97.5 % confidence interval [ CI ] , 0.52 to 0.96 ; P=0.01 ) , and the hazard ratio for the 10-mg macitentan dose as compared with placebo was 0.55 ( 97.5 % CI , 0.39 to 0.76 ; P<0.001 ) . Worsening of pulmonary arterial hypertension was the most frequent primary end-point event . The effect of macitentan on this end point was observed regardless of whether the patient was receiving therapy for pulmonary arterial hypertension at baseline . Adverse events more frequently associated with macitentan than with placebo were headache , nasopharyngitis , and anemia . CONCLUSIONS Macitentan significantly reduced morbidity and mortality among patients with pulmonary arterial hypertension in this event-driven study . ( Funded by Actelion Pharmaceuticals ; SERAPHIN Clinical Trials.gov number , NCT00660179 . )",
"Our previous studies have shown that bone marrow mesenchymal stem cells ( BMSCs ) can inhibit the progression of pulmonary artery hypertension ( PAH ) in the monocrotaline ( MCT ) model in the short term . The aim of this study was to further investigate the long-term effect of BMSCs on PAH and to explore the mechanism of the protective effect including the pulmonary vascular remodeling and cell differentiation . PAH model was established by subcutaneous injection of 50 mg/kg MCT as previously study . Postoperatively , the animals were r and omly divided into three groups ( n = 10 in each group ) : control , PAH group , and BMSCs implantation group . Six months after injection , immunology and immunohistochemistry analysis indicated the MCT-induced intima-media thickness in muscular arteries was reduced ( P < 0.05 ) ; the area of collagen fibers in lung tissue was lower ( P < 0.05 ) , and the proliferating cell nuclear antigen level in pulmonary artery smooth muscle cells was decreased ( P < 0.05 ) . Immunofluorescence showed that the cells have the ability to differentiate between von Willebr and factor and vascular endothelial growth factor . Six months after intravenous injection , BMSCs could significantly improve pulmonary function by inhibiting the ventricular remodeling and the effect of cell differentiation",
"Experimental data suggest that transplantation of EPCs attenuates monocrotaline-induced pulmonary hypertension in rats and dogs . In addition , our previous studies suggested that autologous EPC transplantation was feasible , safe , and might have beneficial effects on exercise capacity and pulmonary hemodynamics in adults with IPAH . Thus , we hypothesized that transplantation of EPCs would improve exercise capacity and pulmonary hemodynamics in children with IPAH . Thirteen children with IPAH received intravenous infusion of autologous EPCs . The right-sided heart catheterization and 6-MWD test were performed at baseline and at the time of 12 wk after cell infusion . At the time of 12 wk , mPAP decreased by 6.4 mmHg from 70.3 + /- 19.0 to 63.9 + /- 19.3 mmHg ( p = 0.015 ) . PVR decreased by approximately 19 % from 1118 + /- 537 to 906 + /- 377 dyn s/cm(5 ) ( p = 0.047 ) . CO increased from 3.39 + /- 0.79 to 3.85 + /- 0.42 L/min ( p = 0.048 ) . The 6-MWD increased by 39 m from 359 + /- 82 to 399 + /- 74 m ( p = 0.012 ) . NYHA functional class also improved . There were no severe adverse events with cell infusion . The small pilot study suggested that intravenous infusion of autologous EPCs was feasible , safe , and associated with significant improvements in exercise capacity , NYHA functional class , and pulmonary hemodynamics in children with IPAH . Confirmation of these results in a r and omized controlled trial are essential",
"BACKGROUND Uncontrolled studies suggested that aerosolized iloprost , a stable analogue of prostacyclin , causes selective pulmonary vasodilatation and improves hemodynamics and exercise capacity in patients with pulmonary hypertension . METHODS We compared repeated daily inhalations of 2.5 or 5.0 microg of iloprost ( six or nine times per day ; median inhaled dose , 30 microg per day ) with inhalation of placebo . A total of 203 patients with selected forms of severe pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension ( New York Heart Association [ NYHA ] functional class III or IV ) were included . The primary end point was met if , after week 12 , the NYHA class and distance walked in six minutes were improved by at least one class and at least 10 percent , respectively , in the absence of clinical deterioration according to predefined criteria and death . RESULTS The combined clinical end point was met by 16.8 percent of the patients receiving iloprost , as compared with 4.9 percent of the patients receiving placebo ( P=0.007 ) . There were increases in the distance walked in six minutes of 36.4 m in the iloprost group as a whole ( P=0.004 ) and of 58.8 m in the subgroup of patients with primary pulmonary hypertension . Overall , 4.0 percent of patients in the iloprost group ( including one who died ) and 13.7 percent of those in the placebo group ( including four who died ) did not complete the study ( P=0.024 ) ; the most common reason for withdrawal was clinical deterioration . As compared with base-line values , hemodynamic values were significantly improved at 12 weeks when measured after iloprost inhalation ( P<0.001 ) , were largely unchanged when measured before iloprost inhalation , and were significantly worse in the placebo group . Further significant beneficial effects of iloprost treatment included an improvement in the NYHA class ( P=0.03 ) , dyspnea ( P=0.015 ) , and quality of life ( P=0.026 ) . Syncope occurred with similar frequency in the two groups but was more frequently rated as serious in the iloprost group , although this adverse effect was not associated with clinical deterioration . CONCLUSIONS Inhaled iloprost is an effective therapy for patients with severe pulmonary hypertension",
"BACKGROUND High pulmonary vascular resistance ( PVR ) may be a risk factor for early and late mortality in both Glen shunt and Fontan operation patients . Furthermore , PVR may increase long after the Fontan operation . Whether pulmonary vasodilators such as phosphodiesterase 5 inhibitors can decrease PVR in patients with single ventricular physiology remains undetermined . METHODS AND RESULTS This was a prospect i ve , multicenter study . Patients with single ventricular physiology who have a PVR index higher than 2.5 Wood units \\u00b7 \\u33a1 ( WU ) were enrolled . Cardiac catheterization was performed before and after administration of sildenafil in all patients . After the Fontan operation , a six minute walk test ( 6MWT ) was also performed . A total of 42 patients were enrolled . PVR was significantly decreased in each stage of single ventricular physiology after sildenafil administration : from 4.3\\u00b11.5WU to 2.1\\u00b10.6WU ( p<0.01 ) in patients before a Glenn shunt , from 3.2\\u00b10.5WU to 1.6\\u00b10.6WU ( p<0.001 ) in patients after a Glenn shunt , and from 3.9\\u00b11.7WU to 2.3\\u00b10.8WU ( p<0.001 ) in patients after Fontan . In patients after Fontan , the 6MWT increased from 416\\u00b174 m to 485\\u00b172 m ( p<0.01 ) , and NYHA functional class improved significantly ( p<0.05 ) after sildenafil administration . No major side effects were observed in any patients . CONCLUSIONS Sildenafil reduced PVR in patients with single ventricle physiology . Sildenafil increased exercise capacity and improved NYHA functional class in patients after a Fontan operation . This implies that pulmonary vasodilation is a potential therapeutic target in selected patients with elevated PVR with single ventricle physiology . Long-term clinical significance warrants further study",
"OBJECTIVES The purpose of this study was to examine the efficacy and safety of four doses of ambrisentan , an oral endothelin type A receptor-selective antagonist , in patients with pulmonary arterial hypertension ( PAH ) . BACKGROUND Pulmonary arterial hypertension is a life-threatening and progressive disease with limited treatment options . Endothelin is a vasoconstrictor and smooth muscle cell mitogen that plays a critical role in the pathogenesis and progression of PAH . METHODS In this double-blind , dose-ranging study , 64 patients with idiopathic PAH or PAH associated with collagen vascular disease , anorexigen use , or human immunodeficiency virus infection were r and omized to receive 1 , 2.5 , 5 , or 10 mg of ambrisentan once daily for 12 weeks followed by 12 weeks of open-label ambrisentan . The primary end point was an improvement from baseline in 6-min walk distance ( 6MWD ) ; secondary end points included Borg dyspnea index , World Health Organization ( WHO ) functional class , a subject global assessment , and cardiopulmonary hemodynamics . RESULTS At 12 weeks , ambrisentan increased 6MWD ( + 36.1 m , p < 0.0001 ) with similar and statistically significant increases for each dose group ( range , + 33.9 to + 38.1 m ) . Improvements were also observed in Borg dyspnea index , WHO functional class , subject global assessment , mean pulmonary arterial pressure ( -5.2 mm Hg , p < 0.0001 ) , and cardiac index ( + 0.33 l/min/m2 , p < 0.0008 ) . Adverse events were mild and unrelated to dose , including the incidence of elevated serum aminotransferase concentrations > 3 times the upper limit of normal ( 3.1 % ) . CONCLUSIONS Ambrisentan appears to improve exercise capacity , symptoms , and hemodynamics in patients with PAH . The incidence and severity of liver enzyme abnormalities appear to be low",
"UNLABELLED Pulmonary arterial hypertension ( PAH ) is characterized by functional and structural changes in the pulmonary vasculature , and despite the drug treatment that made significant progress , the prognosis of patients with advanced PH remains extremely poor . In the present study , we investigated the early effect of bone marrow mesenchymal stem cells ( BMSCs ) on experimental high blood flow-induced PAH model rats and discussed the mechanism . BMSCs were isolated , cultured from bone marrow of Sprague-Dawley ( SD ) rat . The animal model of PAH was created by surgical methods to produce a left-to-right shunt . Following the successful establishment of the PAH model , rats were r and omly assigned to three groups ( n=20 in each group ) : sham group ( control ) , PAH group , and BMSC group ( received a sublingual vein injection of 1 - 5 \\u00d7 10(6 ) BMSCs ) . Two weeks after the administration , BMSCs significantly reduced the vascular remodeling , improved the hemodynamic data , and deceased the right ventricle weight ratio to left ventricular plus septal weight ( RV/LV+S ) ( P<0.05 ) . Real-time reverse transcription-polymerase chain reaction ( RT-PCR ) and immunohistochemistry analysis results indicated that the inflammation factors such as interleukin-1\\u03b2 ( IL-1\\u03b2 ) , IL-6 , and tumor necrosis factor-\\u03b1 ( TNF-\\u03b1 ) were reduced ( P<0.05 ) ; the expression of matrix metallo proteinase-9 ( MMP-9 ) was lower ( P<0.05 ) ; vascular endothelial growth factor ( VEGF ) was higher in BMSC group than those in PAH group ( P<0.05 ) . CONCLUSION Sublingual vein injection of BMSCs for 2 weeks , significantly improved the lung and heart injury caused by left-to-right shunt-induced PAH ; decreased pulmonary vascular remodeling and inflammation ; and enhanced angiogenesis",
"Pulmonary arterial hypertension is a life-threatening disease for which continuous intravenous prostacyclin has proven to be effective . However , this treatment requires a permanent central venous catheter with the associated risk of serious complications such as sepsis , thromboembolism , or syncope . Treprostinil , a stable prostacyclin analogue , can be administered by a continuous subcutaneous infusion , avoiding these risks . We conducted a 12-week , double-blind , placebo-controlled multicenter trial in 470 patients with pulmonary arterial hypertension , either primary or associated with connective tissue disease or congenital systemic-to-pulmonary shunts . Exercise capacity improved with treprostinil and was unchanged with placebo ; the between treatment group difference in median six-minute walking distance was 16 m ( p = 0.006 ) . Improvement in exercise capacity was greater in the sicker patients and was dose-related , but independent of disease etiology . Concomitantly , treprostinil significantly improved indices of dyspnea , signs and symptoms of pulmonary hypertension , and hemodynamics . The most common side effect attributed to treprostinil was infusion site pain ( 85 % ) leading to premature discontinuation from the study in 8 % of patients . Three patients in the treprostinil treatment group presented with an episode of gastrointestinal hemorrhage . We conclude that chronic subcutaneous infusion of treprostinil is an effective treatment with an acceptable safety profile in patients with pulmonary arterial hypertension",
"BACKGROUND Endothelin 1 , a powerful endogenous vasoconstrictor and mitogen , might be a cause of pulmonary hypertension . We describe the efficacy and safety of bosentan , a dual endothelin-receptor antagonist that can be taken orally , in patients with severe pulmonary hypertension . METHODS In this double-blind , placebo-controlled study , 32 patients with pulmonary hypertension ( primary or associated with scleroderma ) were r and omly assigned to bosentan ( 62.5 mg taken twice daily for 4 weeks then 125 mg twice daily ) or placebo for a minimum of 12 weeks . The primary endpoint was change in exercise capacity . Secondary endpoints included changes in cardiopulmonary haemodynamics , Borg dyspnoea index , WHO functional class , and withdrawal due to clinical worsening . Analysis was by intention to treat . FINDINGS In patients given bosentan , the distance walked in 6 min improved by 70 m at 12 weeks compared with baseline , whereas it worsened by 6 m in those on placebo ( difference 76 m [ 95 % CI 12 - 139 ] , p=0.021 ) . The improvement was maintained for at least 20 weeks . The cardiac index was 1.0 L min(-1 ) m(-2 ) ( 95 % CI 0.6 - 1.4 , p<0.0001 ) greater in patients given bosentan than in those given placebo . Pulmonary vascular resistance decreased by 223 dyn s cm(-)(5 ) with bosentan , but increased by 191 dyn s cm(-5 ) with placebo ( difference -415 [ -608 to -221 ] , p=0.0002 ) . Patients given bosentan had a reduced Borg dyspnoea index and an improved WHO functional class . All three withdrawals from clinical worsening were in the placebo group ( p=0.033 ) . The number and nature of adverse events did not differ between the two groups . INTERPRETATION Bosentan increases exercise capacity and improves haemodynamics in patients with pulmonary hypertension , suggesting that endothelin has an important role in pulmonary hypertension",
"Background Systematic Review s ( SRs ) of experimental animal studies are not yet common practice , but awareness of the merits of conducting such SRs is steadily increasing . As animal intervention studies differ from r and omized clinical trials ( RCT ) in many aspects , the methodology for SRs of clinical trials needs to be adapted and optimized for animal intervention studies . The Cochrane Collaboration developed a Risk of Bias ( RoB ) tool to establish consistency and avoid discrepancies in assessing the method ological quality of RCTs . A similar initiative is warranted in the field of animal experimentation . Methods We provide an RoB tool for animal intervention studies ( SYRCLE \\u2019s RoB tool ) . This tool is based on the Cochrane RoB tool and has been adjusted for aspects of bias that play a specific role in animal intervention studies . To enhance transparency and applicability , we formulated signalling questions to facilitate judgment . Results The result ing RoB tool for animal studies contains 10 entries . These entries are related to selection bias , performance bias , detection bias , attrition bias , reporting bias and other biases . Half these items are in agreement with the items in the Cochrane RoB tool . Most of the variations between the two tools are due to differences in design between RCTs and animal studies . Shortcomings in , or unfamiliarity with , specific aspects of experimental design of animal studies compared to clinical studies also play a role . Conclusions SYRCLE \\u2019s RoB tool is an adapted version of the Cochrane RoB tool . Widespread adoption and implementation of this tool will facilitate and improve critical appraisal of evidence from animal studies . This may subsequently enhance the efficiency of translating animal research into clinical practice and increase awareness of the necessity of improving the method ological quality of animal studies"
],
"target": "Conclusions SC therapy is effective for PAH in pre clinical studies .\\nThese results may help to st and ardise pre clinical animal studies and provide a theoretical basis for clinical trial design in the future .",
"background": "Background Despite significant progress in drug treatment , the prognosis of patients with advanced pulmonary arterial hypertension ( PAH ) remains extremely poor .\\nMany pre clinical studies have reported the efficacy of stem cell ( SC ) therapy for PAH ; however , this approach remains controversial .\\nThe aim of this systematic review and meta- analysis is to assess the potential efficacy of SC therapy for PAH .",
"reviews_info": "Background Despite significant progress in drug treatment , the prognosis of patients with advanced pulmonary arterial hypertension ( PAH ) remains extremely poor .\\nMany pre clinical studies have reported the efficacy of stem cell ( SC ) therapy for PAH ; however , this approach remains controversial .\\nThe aim of this systematic review and meta- analysis is to assess the potential efficacy of SC therapy for PAH ."
}
```
__Cochrane__
```json
{
"review_id": "CD007697",
"pmid": [
"16394043"
],
"title": [
"Aggressive surgical effort and improved survival in advanced-stage ovarian cancer."
],
"abstract": [
"Residual disease after initial surgery for ovarian cancer is the strongest prognostic factor for survival. However, the extent of surgical resection required to achieve optimal cytoreduction is controversial. Our goal was to estimate the effect of aggressive surgical resection on ovarian cancer patient survival.\\n A retrospective cohort study of consecutive patients with International Federation of Gynecology and Obstetrics stage IIIC ovarian cancer undergoing primary surgery was conducted between January 1, 1994, and December 31, 1998. The main outcome measures were residual disease after cytoreduction, frequency of radical surgical resection, and 5-year disease-specific survival.\\n The study comprised 194 patients, including 144 with carcinomatosis. The mean patient age and follow-up time were 64.4 and 3.5 years, respectively. After surgery, 131 (67.5%) of the 194 patients had less than 1 cm of residual disease (definition of optimal cytoreduction). Considering all patients, residual disease was the only independent predictor of survival; the need to perform radical procedures to achieve optimal cytoreduction was not associated with a decrease in survival. For the subgroup of patients with carcinomatosis, residual disease and the performance of radical surgical procedures were the only independent predictors. Disease-specific survival was markedly improved for patients with carcinomatosis operated on by surgeons who most frequently used radical procedures compared with those least likely to use radical procedures (44% versus 17%, P < .001).\\n Overall, residual disease was the only independent predictor of survival. Minimizing residual disease through aggressive surgical resection was beneficial, especially in patients with carcinomatosis.\\n II-2."
],
"target": "We found only low quality evidence comparing ultra-radical and standard surgery in women with advanced ovarian cancer and carcinomatosis. The evidence suggested that ultra-radical surgery may result in better survival.\\u00a0 It was unclear whether there were any differences in progression-free survival, QoL and morbidity between the two groups. The cost-effectiveness of this intervention has not been investigated. We are, therefore, unable to reach definite conclusions about the relative benefits and adverse effects of the two types of surgery.\\nIn order to determine the role of ultra-radical surgery in the management of advanced stage ovarian cancer, a sufficiently powered randomised controlled trial comparing ultra-radical and standard surgery or well-designed non-randomised studies would be required."
}
```
### Data Fields
__MS^2__
- `"review_id"`: The PubMed ID of the review.
- `"pmid"`: The PubMed IDs of the included studies.
- `"title"`: The titles of the included studies.
- `"abstract"`: The abstracts of the included studies.
- `"target"`: The conclusions, taken from the abstract of the review, that serve as the summarization target.
- `"background"`: A description of the reviews objective.
__Cochrane__
- `"review_id"`: The PubMed ID of the review.
- `"pmid"`: The PubMed IDs of the included studies.
- `"title"`: The titles of the included studies.
- `"abstract"`: The abstracts of the included studies.
- `"target"`: The conclusions, taken from the abstract of the review, that serve as the summarization target.
### Data Splits
Each dataset is split into training, validation and test partitions
__MS^2__
| train | validation | test |
|------:|-----------:|-----:|
| 14188 | 2021 | 1667 |
__Cochrane__
| train | validation | test |
|------:|-----------:|-----:|
| 3752 | 470 | 470 |
## Dataset Creation
Please refer to the following papers for details about dataset curation:
[MSˆ2: A Dataset for Multi-Document Summarization of Medical Studies](https://aclanthology.org/2021.emnlp-main.594.pdf)
[Generating (Factual?) Narrative Summaries of RCTs: Experiments with Neural Multi-Document Summarization](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8378607/)
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
Licensing information can be found [here](https://github.com/allenai/mslr-shared-task/blob/main/LICENSE).
### Citation Information
**DeYoung, Jay, Iz Beltagy, Madeleine van Zuylen, Bailey Kuehl and Lucy Lu Wang. "MS2: A Dataset for Multi-Document Summarization of Medical Studies." EMNLP (2021).**
```bibtex
@inproceedings{DeYoung2021MS2MS,
title={MSˆ2: Multi-Document Summarization of Medical Studies},
author={Jay DeYoung and Iz Beltagy and Madeleine van Zuylen and Bailey Kuehl and Lucy Lu Wang},
booktitle={EMNLP},
year={2021}
}
```
**Byron C. Wallace, Sayantani Saha, Frank Soboczenski, and Iain James Marshall. (2020). "Generating (factual?) narrative summaries of RCTs: Experiments with neural multi-document summarization." AMIA Annual Symposium.**
```bibtex
@article{Wallace2020GeneratingN,
title={Generating (Factual?) Narrative Summaries of RCTs: Experiments with Neural Multi-Document Summarization},
author={Byron C. Wallace and Sayantani Saha and Frank Soboczenski and Iain James Marshall},
journal={AMIA Annual Symposium},
year={2020},
volume={abs/2008.11293}
}
``` |
false |
# Dataset Card for [Dataset Name]
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Bianet](http://opus.nlpl.eu/Bianet.php)
- **Repository:**
- **Paper:** [Ataman, D. (2018) Bianet: A Parallel News Corpus in Turkish, Kurdish and English. In Proceedings of the LREC 2018 Workshop MLP-Moment. pp. 14-17. pdf](http://lrec-conf.org/workshops/lrec2018/W19/pdf/6_W19.pdf)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
A parallel news corpus in Turkish, Kurdish and English;
Bianet collects 3,214 Turkish articles with their sentence-aligned Kurdish or English translations from the Bianet online newspaper.
3 languages, 3 bitexts
total number of files: 6
total number of tokens: 2.25M
total number of sentence fragments: 0.14M
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
CC-BY-SA-4.0
### Citation Information
@InProceedings{ATAMAN18.6,
author = {Duygu Ataman},
title = {Bianet: A Parallel News Corpus in Turkish, Kurdish and English},
booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)},
year = {2018},
month = {may},
date = {7-12},
location = {Miyazaki, Japan},
editor = {Jinhua Du and Mihael Arcan and Qun Liu and Hitoshi Isahara},
publisher = {European Language Resources Association (ELRA)},
address = {Paris, France},
isbn = {979-10-95546-15-3},
language = {english}
}
### Contributions
Thanks to [@param087](https://github.com/param087) for adding this dataset. |
false |
# Dataset Card for OrangeSum
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** [OrangeSum repository](https://github.com/Tixierae/OrangeSum)
- **Paper:** [BARThez: a Skilled Pretrained French Sequence-to-Sequence Model](https://arxiv.org/abs/2010.12321)
- **Point of Contact:** [Antoine J.-P. Tixier](Antoine.Tixier-1@colorado.edu)
### Dataset Summary
The OrangeSum dataset was inspired by the XSum dataset. It was created by scraping the "Orange Actu" website: https://actu.orange.fr/. Orange S.A. is a large French multinational telecommunications corporation, with 266M customers worldwide. Scraped pages cover almost a decade from Feb 2011 to Sep 2020. They belong to five main categories: France, world, politics, automotive, and society. The society category is itself divided into 8 subcategories: health, environment, people, culture, media, high-tech, unsual ("insolite" in French), and miscellaneous.
Each article featured a single-sentence title as well as a very brief abstract, both professionally written by the author of the article. These two fields were extracted from each page, thus creating two summarization tasks: OrangeSum Title and OrangeSum Abstract.
### Supported Tasks and Leaderboards
**Tasks:** OrangeSum Title and OrangeSum Abstract.
To this day, there is no Leaderboard for this dataset.
### Languages
The text in the dataset is in French.
## Dataset Structure
### Data Instances
A data instance consists of a news article and a summary. The summary can be a short abstract or a title depending on the configuration.
Example:
**Document:** Le temps sera pluvieux sur huit départements de la France ces prochaines heures : outre les trois départements bretons placés en vigilance orange jeudi matin, cinq autres départements du sud du Massif Central ont été à leur tour placés en alerte orange pluie et inondation. Il s'agit de l'Aveyron, du Cantal, du Gard, de la Lozère, et de la Haute-Loire. Sur l'ensemble de l'épisode, les cumuls de pluies attendus en Bretagne sont compris entre 40 et 60 mm en 24 heures et peuvent atteindre localement les 70 mm en 24 heures.Par la suite, la dégradation qui va se mettre en place cette nuit sur le Languedoc et le sud du Massif Central va donner sur l'Aveyron une première salve intense de pluie. Des cumuls entre 70 et 100 mm voir 120 mm localement sont attendus sur une durée de 24 heures. Sur le relief des Cévennes on attend de 150 à 200 mm, voire 250 mm très ponctuellement sur l'ouest du Gard et l'est de la Lozère. Cet épisode va s'estomper dans la soirée avec le décalage des orages vers les régions plus au nord. Un aspect orageux se mêlera à ces précipitations, avec de la grêle possible, des rafales de vent et une forte activité électrique.
**Abstract:** Outre les trois départements bretons, cinq autres départements du centre de la France ont été placés en vigilance orange pluie-inondation.
**Title:** Pluie-inondations : 8 départements en alerte orange.
### Data Fields
`text`: the document to be summarized. \
`summary`: the summary of the source document.
### Data Splits
The data is split into a training, validation and test in both configuration.
| | train | validation | test |
|----------|------:|-----------:|-----:|
| Abstract | 21400 | 1500 | 1500 |
| Title | 30658 | 1500 | 1500 |
## Dataset Creation
### Curation Rationale
The goal here was to create a French equivalent of the recently introduced [XSum](https://github.com/EdinburghNLP/XSum/tree/master/XSum-Dataset) dataset. Unlike the historical summarization datasets, CNN, DailyMail, and NY Times, which favor extractive strategies, XSum, as well as OrangeSum require the models to display a high degree of abstractivity to perform well. The summaries in OrangeSum are not catchy headlines, but rather capture the gist of the articles.
### Source Data
#### Initial Data Collection and Normalization
Each article features a single-sentence title as well as a very brief abstract. Extracting these two fields from each news article page, creates two summarization tasks: OrangeSum Title and OrangeSum Abstract. As a post-processing step, all empty articles and those whose summaries were shorter than 5 words were removed. For OrangeSum Abstract, the top 10% articles in terms of proportion of novel unigrams in the abstracts were removed, as it was observed that such abstracts tend to be introductions rather than real abstracts. This corresponded to a threshold of 57% novel unigrams. For both OrangeSum Title and OrangeSum Abstract, 1500 pairs for testing and 1500 for validation are set aside, and all the remaining ones are used for training.
#### Who are the source language producers?
The authors of the artiles.
### Annotations
#### Annotation process
The smmaries are professionally written by the author of the articles.
#### Who are the annotators?
The authors of the artiles.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
The dataset was initially created by Antoine J.-P. Tixier.
### Licensing Information
[More Information Needed]
### Citation Information
```
@article{eddine2020barthez,
title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},
author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},
journal={arXiv preprint arXiv:2010.12321},
year={2020}
}
```
### Contributions
Thanks to [@moussaKam](https://github.com/moussaKam) for adding this dataset. |
true |
# Dataset Card for Tweets Hate Speech Detection
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Home](https://github.com/sharmaroshan/Twitter-Sentiment-Analysis)
- **Repository:** [Repo](https://github.com/sharmaroshan/Twitter-Sentiment-Analysis/blob/master/train_tweet.csv)
- **Paper:**
- **Leaderboard:**
- **Point of Contact:** [Darshan Gandhi](darshangandhi1151@gmail.com)
### Dataset Summary
The objective of this task is to detect hate speech in tweets. For the sake of simplicity, we say a tweet contains hate speech if it has a racist or sexist sentiment associated with it. So, the task is to classify racist or sexist tweets from other tweets.
Formally, given a training sample of tweets and labels, where label ‘1’ denotes the tweet is racist/sexist and label ‘0’ denotes the tweet is not racist/sexist, your objective is to predict the labels on the given test dataset.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
The tweets are primarily in English Language.
## Dataset Structure
### Data Instances
The dataset contains a label denoting is the tweet a hate speech or not
```
{'label': 0, # not a hate speech
'tweet': ' @user when a father is dysfunctional and is so selfish he drags his kids into his dysfunction. #run'}
```
### Data Fields
* label : 1 - it is a hate speech, 0 - not a hate speech.
* tweet: content of the tweet as a string.
### Data Splits
The data contains training data with :31962 entries
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
Crowdsourced from tweets of users
#### Who are the source language producers?
Cwodsourced from twitter
### Annotations
#### Annotation process
The data has been precprocessed and a model has been trained to assign the relevant label to the tweet
#### Who are the annotators?
The data has been provided by Roshan Sharma
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
With the help of this dataset, one can understand more about the human sentiments and also analye the situations when a particular person intends to make use of hatred/racist comments
### Discussion of Biases
The data could be cleaned up further for additional purposes such as applying a better feature extraction techniques
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Roshan Sharma
### Licensing Information
[Information](https://github.com/sharmaroshan/Twitter-Sentiment-Analysis/blob/master/LICENSE)
### Citation Information
[Citation](https://github.com/sharmaroshan/Twitter-Sentiment-Analysis/blob/master/CONTRIBUTING.md)
### Contributions
Thanks to [@darshan-gandhi](https://github.com/darshan-gandhi) for adding this dataset. |
true |
# Dataset Card for One Million Posts Corpus
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://ofai.github.io/million-post-corpus/
- **Repository:** https://github.com/OFAI/million-post-corpus
- **Paper:** https://dl.acm.org/doi/10.1145/3077136.3080711
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
The “One Million Posts” corpus is an annotated data set consisting of user comments posted to an Austrian newspaper website (in German language).
DER STANDARD is an Austrian daily broadsheet newspaper. On the newspaper’s website, there is a discussion section below each news article where readers engage in online discussions. The data set contains a selection of user posts from the 12 month time span from 2015-06-01 to 2016-05-31. There are 11,773 labeled and 1,000,000 unlabeled posts in the data set. The labeled posts were annotated by professional forum moderators employed by the newspaper.
The data set contains the following data for each post:
* Post ID
* Article ID
* Headline (max. 250 characters)
* Main Body (max. 750 characters)
* User ID (the user names used by the website have been re-mapped to new numeric IDs)
* Time stamp
* Parent post (replies give rise to tree-like discussion thread structures)
* Status (online or deleted by a moderator)
* Number of positive votes by other community members
* Number of negative votes by other community members
For each article, the data set contains the following data:
* Article ID
* Publishing date
* Topic Path (e.g.: Newsroom / Sports / Motorsports / Formula 1)
* Title
* Body
Detailed descriptions of the post selection and annotation procedures are given in the paper.
#### Annotated Categories
Potentially undesirable content:
* Sentiment (negative/neutral/positive)
An important goal is to detect changes in the prevalent sentiment in a discussion, e.g., the location within the fora and the point in time where a turn from positive/neutral sentiment to negative sentiment takes place.
* Off-Topic (yes/no)
Posts which digress too far from the topic of the corresponding article.
* Inappropriate (yes/no)
Swearwords, suggestive and obscene language, insults, threats etc.
* Discriminating (yes/no)
Racist, sexist, misogynistic, homophobic, antisemitic and other misanthropic content.
Neutral content that requires a reaction:
* Feedback (yes/no)
Sometimes users ask questions or give feedback to the author of the article or the newspaper in general, which may require a reply/reaction.
Potentially desirable content:
* Personal Stories (yes/no)
In certain fora, users are encouraged to share their personal stories, experiences, anecdotes etc. regarding the respective topic.
* Arguments Used (yes/no)
It is desirable for users to back their statements with rational argumentation, reasoning and sources.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
Austrian German
## Dataset Structure
### Data Instances
An example from the `posts_labeled` config:
```json
{
"ID_Post": "79",
"ID_Parent_Post": "",
"ID_Article": "1",
"ID_User": "12071",
"CreatedAt": "2015-06-01 08:58:32.363",
"Status": "online",
"Headline": "",
"Body": "ich kann keinen hinweis finden, wo man sich hinwenden muss, sollte man als abonnent des standard, die zeitung nicht bekommt, ist dass bewusst so arrangiert?",
"PositiveVotes": 0,
"NegativeVotes": 0,
"Category": 5,
"Value": 1,
"Fold": 1
}
```
An example from the `posts_unlabeled` config:
```json
{
"ID_Post": "51",
"ID_Parent_Post": "",
"ID_Article": "1",
"ID_User": "11125",
"CreatedAt": "2011-05-15 08:37:11.313",
"Status": "online",
"Headline": "Ich würde es sehr begrüßen, wenn",
"Body": "Antworten erst beim Erscheinen als e-Mail dem Poster zugestellt würden.\r\n\r\nEs gibt User, die ihre Kommentare sofort nach Mail-Eingang irgendwo hinposten. Dadurch wird \r\n1. vor allem für andere Unser die Lesbarkeit wesentlich beeinträchtigt,\r\n2. kann das Post verdreht wiedergegeben werden,\r\n3. man ist immer wieder gezwungen die Antwort richtig zu stellen.\r\n\r\nPrivatfehden von Usern sollten, wenn schon zugelassen, für alle User nachvollziehbar sein.\r\n\r\nDanke!",
"PositiveVotes": 1,
"NegativeVotes": 0
}
```
An example from the `articles` config:
```json
{
"ID_Article": "41",
"Path": "Newsroom/Wirtschaft/Wirtschaftpolitik/Energiemarkt",
"publishingDate": "2015-06-01 12:39:35.00",
"Title": "Öl- und Gas-Riesen fordern weltweite CO2-Preise",
"Body": '<div class="section" id="content-main" itemprop="articleBody"><div class="copytext"><h2 itemprop="description">Brief von BP, Total, Shell, Statoil, BG Group und Eni unterzeichnet</h2><p>Paris/London/La Defense - Sechs große Öl- und Gaskonzerne haben mit Blick auf die Verhandlungen über einen neuen Welt-Klimavertrag ein globales Preissystem für CO2-Emissionen gefordert. Wenn der Ausstoß von CO2 Geld kostet, sei dies ein Anreiz für die Nutzung von Erdgas statt Kohle, mehr Energieeffizienz und Investitionen zur Vermeidung des Treibhausgases, heißt es in einem am Montag veröffentlichten Brief.</p>\n<p>Das Schreiben ist unterzeichnet von BP, Total, Shell, Statoil, BG Group und Eni. Die Unternehmen versicherten, sie seien bereit, ihren Teil zum Kampf gegen den <a href="/r1937/Klimawandel">Klimawandel</a> beizutragen. Dafür sei aber ein klarer und verlässlicher Politik-Rahmen nötig. (APA, 1.6.2015)</p> </div></div>'
}
```
### Data Fields
The data set contains the following data for each post:
* **ID_Post**: Post ID
* **ID_Parent_Post**: Parent post (replies give rise to tree-like discussion thread structures)
* **ID_Article**: Article ID
* **ID_User**: User ID (the user names used by the website have been re-mapped to new numeric IDs)
* **Headline**: Headline (max. 250 characters)
* **Body**: Main Body (max. 750 characters)
* **CreatedAt**: Time stamp
* **Status**: Status (online or deleted by a moderator)
* **PositiveVotes**: Number of positive votes by other community members
* **NegativeVotes**: Number of negative votes by other community members
Labeled posts also contain:
* **Category**: The category of the annotation, one of: ArgumentsUsed, Discriminating, Inappropriate, OffTopic, PersonalStories, PossiblyFeedback, SentimentNegative, SentimentNeutral, SentimentPositive
* **Value**: either 0 or 1, explicitly indicating whether or not the post has the specified category as a label (i.e. a category of `ArgumentsUsed` with value of `0` means that an annotator explicitly labeled that this post doesn't use arguments, as opposed to the mere absence of a positive label).
* **Fold**: a number between [0-9] from a 10-fold split by the authors
For each article, the data set contains the following data:
* **ID_Article**: Article ID
* **publishingDate**: Publishing date
* **Path**: Topic Path (e.g.: Newsroom / Sports / Motorsports / Formula 1)
* **Title**: Title
* **Body**: Body
### Data Splits
Training split only.
| name | train |
|-----------------|--------:|
| posts_labeled | 40567 |
| posts_unlabeled | 1000000 |
| articles | 12087 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
This data set is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
### Citation Information
```
@InProceedings{Schabus2018,
author = {Dietmar Schabus and Marcin Skowron},
title = {Academic-Industrial Perspective on the Development and Deployment of a Moderation System for a Newspaper Website},
booktitle = {Proceedings of the 11th International Conference on Language Resources and Evaluation (LREC)},
year = {2018},
address = {Miyazaki, Japan},
month = may,
pages = {1602-1605},
abstract = {This paper describes an approach and our experiences from the development, deployment and usability testing of a Natural Language Processing (NLP) and Information Retrieval system that supports the moderation of user comments on a large newspaper website. We highlight some of the differences between industry-oriented and academic research settings and their influence on the decisions made in the data collection and annotation processes, selection of document representation and machine learning methods. We report on classification results, where the problems to solve and the data to work with come from a commercial enterprise. In this context typical for NLP research, we discuss relevant industrial aspects. We believe that the challenges faced as well as the solutions proposed for addressing them can provide insights to others working in a similar setting.},
url = {http://www.lrec-conf.org/proceedings/lrec2018/summaries/8885.html},
}
```
### Contributions
Thanks to [@aseifert](https://github.com/aseifert) for adding this dataset. |
false |
# Dataset Card for SimpleQuestions
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://research.fb.com/downloads/babi/
- **Repository:** https://github.com/fbougares/TSAC
- **Paper:** https://research.fb.com/publications/large-scale-simple-question-answering-with-memory-networks/
- **Leaderboard:** [If the dataset supports an active leaderboard, add link here]()
- **Point of Contact:** [Antoine Borde](abordes@fb.com) [Nicolas Usunie](usunier@fb.com) [Sumit Chopra](spchopra@fb.com), [Jason Weston](jase@fb.com)
### Dataset Summary
[More Information Needed]
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
Here are some examples of questions and facts:
* What American cartoonist is the creator of Andy Lippincott?
Fact: (andy_lippincott, character_created_by, garry_trudeau)
* Which forest is Fires Creek in?
Fact: (fires_creek, containedby, nantahala_national_forest)
* What does Jimmy Neutron do?
Fact: (jimmy_neutron, fictional_character_occupation, inventor)
* What dietary restriction is incompatible with kimchi?
Fact: (kimchi, incompatible_with_dietary_restrictions, veganism)
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
[More Information Needed]
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
[More Information Needed]
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@abhishekkrthakur](https://github.com/abhishekkrthakur) for adding this dataset. |
false |
# Dataset Card for Thai Literature Corpora (TLC)
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://attapol.github.io/tlc.html
- **Leaderboard:** https://www.kaggle.com/c/wisesight-sentiment/
- **Paper:**
- **Leaderboard:**
- **Point of Contact:** Jitkapat Sawatphol, Attapol Rutherford; attapolrutherford at gmail.com
### Dataset Summary
Thai Literature Corpora (TLC): Corpora of machine-ingestible Thai classical literature texts.
It consists of two datasets:
## TLC set
It is texts from [Vajirayana Digital Library](https://vajirayana.org/), stored by chapters and stanzas (non-tokenized).
tlc v.2.0 (6/17/19 : a total of 34 documents, 292,270 lines, 31,790,734 characters)
tlc v.1.0 (6/11/19 : a total of 25 documents, 113,981 lines, 28,775,761 characters)
## TNHC set
It is texts from Thai National Historical Corpus, stored by lines (manually tokenized).
tnhc v.1.0 (6/25/19 : a total of 47 documents, 756,478 lines, 13,361,142 characters)
### Supported Tasks and Leaderboards
Language Modeling, Language Generation
### Languages
Thai
## Dataset Structure
### Data Instances
```
{
"ch_num": "๑",
"title": "กากี กลอนสุภาพ",
"text": [
[
"๏ จักกล่าวอดีตนิทานแต่ปางก่อน\n",
"เมื่อครั้งองค์สมเด็จพระชินวร\tยังสัญจรแสวงหาโพธิญาณ\n",
"เสวยชาติเป็นสกุณาพระยานก\tจึงชักเรื่องชาดกมาบรรหาร\n",
"หวังแสดงแห่งจิตหญิงพาล\tให้ชายชาญรู้เชิงกระสัตรี ฯ\n"
]
}
```
### Data Fields
- `ch_num`: chapter number in Thai Numerals (๑, ๒, ๓, ๔, ๕, ๖, ๗, ๘, ๙, ๑๐, ...)
- `title`: chapter name
- `text`: each item corresponds to one stanzas, each line is a couplet which can be seperated by `\t`
### Data Splits
tlc v.2.0 (6/17/19 : a total of 34 documents, 292,270 lines, 31,790,734 characters)
tlc v.1.0 (6/11/19 : a total of 25 documents, 113,981 lines, 28,775,761 characters)
## TNHC set
It is texts from Thai National Historical Corpus, stored by lines (manually tokenized).
tnhc v.1.0 (6/25/19 : a total of 47 documents, 756,478 lines, 13,361,142 characters)
| | tlc2.0 | tlc1.0 | tnhc |
|-----------|-------|-------|-------|
| # documents | 34 | 25 | 47 |
| # lines | 292,270 | 113,981 | 756,478 |
## Dataset Creation
### Curation Rationale
Originally, the dataset was compiled for the [Thai Poetry Generator](https://github.com/jitkapat/thaipoetrygenerator) at Chulalongkorn university as the Final project for `2209372 Introduction to Computational Linguistics` by [Jitkapat Sawatphol](https://jitkapat.github.io/) (Faculty of Engineering, Chulalongkorn University).
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
There is no personal information.
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Thanks [Jitkapat Sawatphol](https://jitkapat.github.io/) (Faculty of Arts, Chulalongkorn University), and [Attapol Rutherford](https://attapol.github.io/) (Faculty of Arts, Chulalongkorn University)
### Licensing Information
[More Information Needed]
### Citation Information
Please cite the following if you make use of the dataset:
Jitkapat Sawatphol, and Attapol Rutherford. 2019. **Thai Literature Corpora (TLC)**.
BibTeX:
```
@misc{
author={Sawatphol, Jitkapat},
title={Thai Literature Corpora},
year={2019},
howpublished={\\url{https://attapol.github.io/tlc.html}}
}
```
### Contributions
Thanks to [@chameleonTK](https://github.com/chameleonTK) for adding this dataset. |
false |
# Dataset Card for PEC
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** [PEC repository](https://github.com/zhongpeixiang/PEC)
- **Paper:** [Towards Persona-Based Empathetic Conversational Models](https://www.aclweb.org/anthology/2020.emnlp-main.531/)
- **Point of Contact:** [Peixiang Zhong](mailto:zhongpeixiang@gmail.com)
### Dataset Summary
The PEC dataset is an English-language dataset of open-domain conversations gathered from two subreddits on Reddit, i.e., happy and offmychest. PEC has around 350K persona-based empathetic conversations. Each utterance is associated with a speaker, and each speaker has a persona of multiple persona sentences. The conversations in PEC are more empathetic than casual conversations. The conversations in the happy domain are mostly positive, whereas the conversations in the offmychest domain are mostly negative.
### Supported Tasks and Leaderboards
- `dialogue-modeling`, `utterance-retrieval`: this dataset can be used to train a generative or retrieval-based conversational model.
### Languages
English
## Dataset Structure
### Data Instances
A typical data example comprises a list of context utterances, a list of context speakers, a response to the context, the response speaker and the persona of the response speaker.
An example from PEC looks as follows:
```
{'context': ['found out this morning i got a job promotion ! ! !'],
'context_speakers': ['HeWentToJared91'],
'personas': [
"i ca n't stand working in the ugli .",
'i ’ve always liked my eyes except for the fact that they ca n’t shoot lasers',
'i feel really bad about myself as a person right now , and i could really use a hand .',
'i drank a coffee , and it just made me feel even more exhausted .',
'i want a natsuki t shirt',
"i 've dealt with depression in the past .",
'i love red dead 2'],
'response': "you look like a nice person ! we 're proud of you , and i bet you earned that promotion !",
'response_speaker': 'tylock'}
```
### Data Fields
- `context`: a list of strings, each string denotes a context utterance.
- `context_speakers`: a list of strings, each string denotes a speaker.
- `response`: a string denoting the response to the `context`.
- `response_speaker`: a string denoting the speaker of `response`.
- `personas`: a list of strings, each string denotes a persona sentence of `response_speaker`.
### Data Splits
The data is split into a training, validation and test set for each of the three domains. Note that the *all* domain is the concatenation of the *happy* and *offmychest* domains.
| domain | train | validation | test |
|------------|-------:|-----------:|------:|
| happy | 157195 | 19829 | 22730 |
| offmychest | 123968 | 16004 | 15324 |
| all | 281163 | 35833 | 38054 |
## Dataset Creation
### Curation Rationale
PEC was built to provide a testbed for machines to learn persona-based empathetic responding. In our empirical analysis, we found that different personas have different styles of empathetic responding. This dataset can also be used to investigate the link between persona and empathy in human conversations. According to our human assessment, the conversations on the happy and offmychest subreddits are significantly more empathetic than casual conversations.
### Source Data
#### Initial Data Collection and Normalization
The data was obtained via the [pushshift API](https://pushshift.io/using-bigquery-with-reddit-data/) via Google BigQuery.
#### Who are the source language producers?
The language producers are users of the [r/happy](https://www.reddit.com/r/happy/), and [r/offmychest](https://www.reddit.com/r/offmychest/) subreddits between 2012 and 2020. No further demographic information was available from the data source.
### Annotations
#### Annotation process
The dataset does not contain any additional annotations.
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
The dataset includes the speaker IDs of users on *happy* and *offmychest* subreddits.
## Considerations for Using the Data
### Social Impact of Dataset
The purpose of this dataset is to help develop more personalised and empathetic conversational systems, which is an important milestone towards truly human-like conversational agents.
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
A small portion of the dataset has the issues of sexism, hate, and harassment. The persona sentences are noisy.
## Additional Information
### Dataset Curators
The dataset was initially created by Peixiang Zhong, Chen Zhang, Hao Wang, Yong Liu, and Chunyan Miao, jointly done at Nanyang Technological University and Alibaba Group.
### Licensing Information
The licensing status of the dataset hinges on the legal status of the [Pushshift.io](https://files.pushshift.io/reddit/) data which is unclear.
### Citation Information
```
@inproceedings{zhong-etal-2020-towards,
title = "Towards Persona-Based Empathetic Conversational Models",
author = "Zhong, Peixiang and
Zhang, Chen and
Wang, Hao and
Liu, Yong and
Miao, Chunyan",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
year = "2020",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.531",
pages = "6556--6566"
}
```
### Contributions
Thanks to [@zhongpeixiang](https://github.com/zhongpeixiang) for adding this dataset. |
false |
# Dataset Card for UMC005 English-Urdu
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** http://ufal.ms.mff.cuni.cz/umc/005-en-ur/
- **Repository:** None
- **Paper:** https://www.researchgate.net/publication/268008206_Word-Order_Issues_in_English-to-Urdu_Statistical_Machine_Translation
- **Leaderboard:** [If the dataset supports an active leaderboard, add link here]()
- **Point of Contact:** Bushra Jawaid and Daniel Zeman
### Dataset Summary
[More Information Needed]
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
[More Information Needed]
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
[More Information Needed]
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@abhishekkrthakur](https://github.com/abhishekkrthakur) for adding this dataset. |
false |
# Dataset Card for the Dog 🐶 vs. Food 🍔 (a.k.a. Dog Food) Dataset
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**: https://github.com/qw2243c/Image-Recognition-Dogs-Fried-Chicken-or-Blueberry-Muffins-
- **Repository:** : https://github.com/qw2243c/Image-Recognition-Dogs-Fried-Chicken-or-Blueberry-Muffins-
- **Paper:** : N/A
- **Leaderboard:**: N/A
- **Point of Contact:**: @sasha
### Dataset Summary
This is a dataset for multiclass image classification, between 'dog', 'chicken', and 'muffin' classes.
The 'dog' class contains images of dogs that look like fried chicken and some that look like images of muffins, while the 'chicken' and 'muffin' classes contains images of (you guessed it) fried chicken and muffins 😋
### Supported Tasks and Leaderboards
TBC
### Languages
The labels are in English (['dog', 'chicken', 'muffin'])
## Dataset Structure
### Data Instances
A sample from the training set is provided below:
```
{
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=300x470 at 0x7F176094EF28>,
'label': 0}
}
```
### Data Fields
- img: A `PIL.JpegImageFile` object containing the 300x470. image. Note that when accessing the image column: `dataset[0]["image"]` the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the `"image"` column, *i.e.* `dataset[0]["image"]` should **always** be preferred over `dataset["image"][0]`
- label: 0-1 with the following correspondence
0 dog
1 food
### Data Splits
Train (1875 images) and Test (625 images)
## Dataset Creation
### Curation Rationale
N/A
### Source Data
#### Initial Data Collection and Normalization
This dataset was taken from the [qw2243c/Image-Recognition-Dogs-Fried-Chicken-or-Blueberry-Muffins?](https://github.com/qw2243c/Image-Recognition-Dogs-Fried-Chicken-or-Blueberry-Muffins-) Github repository and randomly splitting 25% of the data for validation.
### Annotations
#### Annotation process
This data was scraped from the internet and annotated based on the query words.
### Personal and Sensitive Information
N/A
## Considerations for Using the Data
### Social Impact of Dataset
N/A
### Discussion of Biases
This dataset is balanced -- it has an equal number of images of dogs (1000) compared to chicken (1000 and muffin (1000). This should be taken into account when evaluating models.
### Other Known Limitations
N/A
## Additional Information
### Dataset Curators
This dataset was created by @lanceyjt, @yl3829, @wesleytao, @qw2243c and @asyouhaveknown
### Licensing Information
No information is indicated on the original [github repository](https://github.com/qw2243c/Image-Recognition-Dogs-Fried-Chicken-or-Blueberry-Muffins-).
### Citation Information
N/A
### Contributions
Thanks to [@lewtun](https://github.com/lewtun) for adding this dataset.
|
false |
# Dataset Card for GEM/xsum
## Dataset Description
- **Homepage:** n/a
- **Repository:** https://github.com/EdinburghNLP/XSum
- **Paper:** https://www.aclweb.org/anthology/D18-1206
- **Leaderboard:** N/A
- **Point of Contact:** Shashi Narayan
### Link to Main Data Card
You can find the main data card on the [GEM Website](https://gem-benchmark.com/data_cards/xsum).
### Dataset Summary
XSum is an English news summarization dataset where the task is to predict the first sentence of an article from the rest of it.
You can load the dataset via:
```
import datasets
data = datasets.load_dataset('GEM/xsum')
```
The data loader can be found [here](https://huggingface.co/datasets/GEM/xsum).
#### website
n/a
#### paper
[ACL Anthology](https://www.aclweb.org/anthology/D18-1206)
#### authors
Shashi Narayan, Shay B. Cohen, Mirella Lapata (all affiliated with University of Edinburgh at the time of dataset creation)
## Dataset Overview
### Where to find the Data and its Documentation
#### Download
<!-- info: What is the link to where the original dataset is hosted? -->
<!-- scope: telescope -->
[Github](https://github.com/EdinburghNLP/XSum)
#### Paper
<!-- info: What is the link to the paper describing the dataset (open access preferred)? -->
<!-- scope: telescope -->
[ACL Anthology](https://www.aclweb.org/anthology/D18-1206)
#### BibTex
<!-- info: Provide the BibTex-formatted reference for the dataset. Please use the correct published version (ACL anthology, etc.) instead of google scholar created Bibtex. -->
<!-- scope: microscope -->
```
@InProceedings{xsum-emnlp,
author = "Shashi Narayan and Shay B. Cohen and Mirella Lapata",
title = "Don't Give Me the Details, Just the Summary! {T}opic-Aware Convolutional Neural Networks for Extreme Summarization",
booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing ",
year = "2018",
address = "Brussels, Belgium",
}
```
#### Contact Name
<!-- quick -->
<!-- info: If known, provide the name of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
Shashi Narayan
#### Contact Email
<!-- info: If known, provide the email of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
shashinarayan@google.com
#### Has a Leaderboard?
<!-- info: Does the dataset have an active leaderboard? -->
<!-- scope: telescope -->
no
### Languages and Intended Use
#### Multilingual?
<!-- quick -->
<!-- info: Is the dataset multilingual? -->
<!-- scope: telescope -->
no
#### Covered Dialects
<!-- info: What dialects are covered? Are there multiple dialects per language? -->
<!-- scope: periscope -->
Since the source of the dataset are BBC articles, the language is in British English of the variation written by journalists.
#### Covered Languages
<!-- quick -->
<!-- info: What languages/dialects are covered in the dataset? -->
<!-- scope: telescope -->
`English`
#### Whose Language?
<!-- info: Whose language is in the dataset? -->
<!-- scope: periscope -->
Professional journalists
#### License
<!-- quick -->
<!-- info: What is the license of the dataset? -->
<!-- scope: telescope -->
cc-by-sa-4.0: Creative Commons Attribution Share Alike 4.0 International
#### Intended Use
<!-- info: What is the intended use of the dataset? -->
<!-- scope: microscope -->
The dataset is for the task of abstractive summarization in its extreme form, its about summarizing a document in a single sentence. The idea is to create a short, one-sentence news summary answering the question "What is the article about?".
#### Primary Task
<!-- info: What primary task does the dataset support? -->
<!-- scope: telescope -->
Summarization
#### Communicative Goal
<!-- quick -->
<!-- info: Provide a short description of the communicative goal of a model trained for this task on this dataset. -->
<!-- scope: periscope -->
Given a news article, produce a single sentence summary of the content of the article.
### Credit
#### Curation Organization Type(s)
<!-- info: In what kind of organization did the dataset curation happen? -->
<!-- scope: telescope -->
`academic`
#### Curation Organization(s)
<!-- info: Name the organization(s). -->
<!-- scope: periscope -->
University of Edinburgh
#### Dataset Creators
<!-- info: Who created the original dataset? List the people involved in collecting the dataset and their affiliation(s). -->
<!-- scope: microscope -->
Shashi Narayan, Shay B. Cohen, Mirella Lapata (all affiliated with University of Edinburgh at the time of dataset creation)
#### Funding
<!-- info: Who funded the data creation? -->
<!-- scope: microscope -->
European Research Council (Lapata; award number 681760), the European Union under the Horizon 2020 SUMMA project (Narayan, Cohen; grant agreement 688139), and Huawei Technologies (Cohen).
#### Who added the Dataset to GEM?
<!-- info: Who contributed to the data card and adding the dataset to GEM? List the people+affiliations involved in creating this data card and who helped integrate this dataset into GEM. -->
<!-- scope: microscope -->
The original data card was written by Laura Perez-Beltrachini and the data loader by Yacine Jernite. Sebastian Gehrmann migrated the data card to the new format and extended it. The v2 data loader was migrated by Abinaya Mahendiran
### Dataset Structure
#### Data Fields
<!-- info: List and describe the fields present in the dataset. -->
<!-- scope: telescope -->
- `Document`: Input news article.
- `Summary`: One sentence summary of the article.
- `Id`: BBC ID of the article.
#### Reason for Structure
<!-- info: How was the dataset structure determined? -->
<!-- scope: microscope -->
The Document/Summary format is standard for summarization datasets.
#### How were labels chosen?
<!-- info: How were the labels chosen? -->
<!-- scope: microscope -->
The labels are the first sentence of the source article.
#### Example Instance
<!-- info: Provide a JSON formatted example of a typical instance in the dataset. -->
<!-- scope: periscope -->
```
{
'document': 'The researchers have sequenced the genome of a strain of bacterium that causes the virulent infection.\nA survey in 2007 showed that bleeding canker had spread rapidly, with almost half of the two million horse chestnuts displaying symptoms of the disease.\nThe findings have been published in the journal PLoS One.\nA visible symptom of the disease is a lesion on the bark, which oozes a resin on to the trunk or sometimes the branches.\nThe bark underneath the canker is killed, and if cankers manage to go all the way around the trunk then the horse chestnut (Aesculus hippocastanum) will die because it cuts off the food supply. [...]',
'target': "A team of UK scientists hopes to shed light on the mysteries of bleeding canker, a disease that is threatening the nation's horse chestnut trees.",
}
```
#### Data Splits
<!-- info: Describe and name the splits in the dataset if there are more than one. -->
<!-- scope: periscope -->
| Section | Number of Documents |
| ------------- |:-------------:|
| Training | 204,045 |
| Validation | 11,332 |
| Testing | 11,334 |
| Total | 226k |
| Section | number of words| number of sentences |
| ------------- |:-------------:| :-------------:|
| Documents | 431.07 | 19.77 |
| Summary | 23.26 | 1.00 |
#### Splitting Criteria
<!-- info: Describe any criteria for splitting the data, if used. If there are differences between the splits (e.g., if the training annotations are machine-generated and the dev and test ones are created by humans, or if different numbers of annotators contributed to each example), describe them here. -->
<!-- scope: microscope -->
The identifiers in the URLs were used to randomly split the dataset into training (90%, 204,045), validation (5%, 11,332), and test (5%, 11,334) sets.
## Dataset Curation
### Original Curation
#### Original Curation Rationale
<!-- info: Original curation rationale -->
<!-- scope: telescope -->
Comparable datasets are often very extractive which is not a strategy that works for one-sentence summaries. The dataset curators thus created this dataset as a way to evaluate truly abstractive models
#### Communicative Goal
<!-- info: What was the communicative goal? -->
<!-- scope: periscope -->
Same as the communicative goal in GEM: A model should summarize a news article in a single sentence
#### Sourced from Different Sources
<!-- info: Is the dataset aggregated from different data sources? -->
<!-- scope: telescope -->
no
### Language Data
#### How was Language Data Obtained?
<!-- info: How was the language data obtained? -->
<!-- scope: telescope -->
`Found`
#### Where was it found?
<!-- info: If found, where from? -->
<!-- scope: telescope -->
`Single website`
#### Language Producers
<!-- info: What further information do we have on the language producers? -->
<!-- scope: microscope -->
The data was collected from articles between 2010 and 2017. No other information
#### Topics Covered
<!-- info: Does the language in the dataset focus on specific topics? How would you describe them? -->
<!-- scope: periscope -->
The collected articles included the following topics: News, Politics, Sports, Weather, Business, Technology, Science, Health, Family, Education, Entertainment and Arts
The dataset curators also used LDA to gain insight into this question and found that the following were the top keywords associated with each topic:
- **T1**: charge, court, murder, police, arrest, guilty, sentence, boy, bail, space, crown, trial
- **T2**: church, abuse, bishop, child, catholic, gay, pope, school, christian, priest, cardinal
- **T3**: council, people, government, local, housing, home, house, property, city, plan, authority
- **T4**: clinton, party, trump, climate, poll, vote, plaid, election, debate, change, candidate, campaign
- **T5**: country, growth, report, business, export, fall, bank, security, economy, rise, global, inflation
- **T6**: hospital, patient, trust, nhs, people, care, health, service, staff, report, review, system, child
#### Data Validation
<!-- info: Was the text validated by a different worker or a data curator? -->
<!-- scope: telescope -->
not validated
#### Data Preprocessing
<!-- info: How was the text data pre-processed? (Enter N/A if the text was not pre-processed) -->
<!-- scope: microscope -->
The text was extracted from the HTML of the webpage. No further processing was done.
#### Was Data Filtered?
<!-- info: Were text instances selected or filtered? -->
<!-- scope: telescope -->
not filtered
### Structured Annotations
#### Additional Annotations?
<!-- quick -->
<!-- info: Does the dataset have additional annotations for each instance? -->
<!-- scope: telescope -->
none
#### Annotation Service?
<!-- info: Was an annotation service used? -->
<!-- scope: telescope -->
no
### Consent
#### Any Consent Policy?
<!-- info: Was there a consent policy involved when gathering the data? -->
<!-- scope: telescope -->
no
#### Justification for Using the Data
<!-- info: If not, what is the justification for reusing the data? -->
<!-- scope: microscope -->
The copyright license of the data allows reusing it for this purpose.
### Private Identifying Information (PII)
#### Contains PII?
<!-- quick -->
<!-- info: Does the source language data likely contain Personal Identifying Information about the data creators or subjects? -->
<!-- scope: telescope -->
yes/very likely
#### Categories of PII
<!-- info: What categories of PII are present or suspected in the data? -->
<!-- scope: periscope -->
`generic PII`
#### Any PII Identification?
<!-- info: Did the curators use any automatic/manual method to identify PII in the dataset? -->
<!-- scope: periscope -->
no identification
### Maintenance
#### Any Maintenance Plan?
<!-- info: Does the original dataset have a maintenance plan? -->
<!-- scope: telescope -->
no
## Broader Social Context
### Previous Work on the Social Impact of the Dataset
#### Usage of Models based on the Data
<!-- info: Are you aware of cases where models trained on the task featured in this dataset ore related tasks have been used in automated systems? -->
<!-- scope: telescope -->
no
### Impact on Under-Served Communities
#### Addresses needs of underserved Communities?
<!-- info: Does this dataset address the needs of communities that are traditionally underserved in language technology, and particularly language generation technology? Communities may be underserved for exemple because their language, language variety, or social or geographical context is underepresented in NLP and NLG resources (datasets and models). -->
<!-- scope: telescope -->
no
### Discussion of Biases
#### Any Documented Social Biases?
<!-- info: Are there documented social biases in the dataset? Biases in this context are variations in the ways members of different social categories are represented that can have harmful downstream consequences for members of the more disadvantaged group. -->
<!-- scope: telescope -->
unsure
#### Are the Language Producers Representative of the Language?
<!-- info: Does the distribution of language producers in the dataset accurately represent the full distribution of speakers of the language world-wide? If not, how does it differ? -->
<!-- scope: periscope -->
The language and content of the data is focused on news and language in the UK and as such not representative of the speakers world-wide. Existing selection biases of the BBC exist in this dataset.
|
false |
# Dataset Card for BSD100
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage**: https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/bsds/
- **Repository**: https://huggingface.co/datasets/eugenesiow/BSD100
- **Paper**: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=937655
- **Leaderboard**: https://github.com/eugenesiow/super-image#scale-x2
### Dataset Summary
BSD is a dataset used frequently for image denoising and super-resolution. Of the subdatasets, BSD100 is aclassical image dataset having 100 test images proposed by [Martin et al. (2001)](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=937655). The dataset is composed of a large variety of images ranging from natural images to object-specific such as plants, people, food etc. BSD100 is the testing set of the Berkeley segmentation dataset BSD300.
Install with `pip`:
```bash
pip install datasets super-image
```
Evaluate a model with the [`super-image`](https://github.com/eugenesiow/super-image) library:
```python
from datasets import load_dataset
from super_image import EdsrModel
from super_image.data import EvalDataset, EvalMetrics
dataset = load_dataset('eugenesiow/BSD100', 'bicubic_x2', split='validation')
eval_dataset = EvalDataset(dataset)
model = EdsrModel.from_pretrained('eugenesiow/edsr-base', scale=2)
EvalMetrics().evaluate(model, eval_dataset)
```
### Supported Tasks and Leaderboards
The dataset is commonly used for evaluation of the `image-super-resolution` task.
Unofficial [`super-image`](https://github.com/eugenesiow/super-image) leaderboard for:
- [Scale 2](https://github.com/eugenesiow/super-image#scale-x2)
- [Scale 3](https://github.com/eugenesiow/super-image#scale-x3)
- [Scale 4](https://github.com/eugenesiow/super-image#scale-x4)
- [Scale 8](https://github.com/eugenesiow/super-image#scale-x8)
### Languages
Not applicable.
## Dataset Structure
### Data Instances
An example of `validation` for `bicubic_x2` looks as follows.
```
{
"hr": "/.cache/huggingface/datasets/downloads/extracted/BSD100_HR/3096.png",
"lr": "/.cache/huggingface/datasets/downloads/extracted/BSD100_LR_x2/3096.png"
}
```
### Data Fields
The data fields are the same among all splits.
- `hr`: a `string` to the path of the High Resolution (HR) `.png` image.
- `lr`: a `string` to the path of the Low Resolution (LR) `.png` image.
### Data Splits
| name |validation|
|-------|---:|
|bicubic_x2|100|
|bicubic_x3|100|
|bicubic_x4|100|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
No annotations.
#### Who are the annotators?
No annotators.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
- **Original Authors**: [Martin et al. (2001)](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=937655)
### Licensing Information
You are free to download a portion of the dataset for non-commercial research and educational purposes.
In exchange, we request only that you make available to us the results of running your segmentation or
boundary detection algorithm on the test set as described below. Work based on the dataset should cite
the [Martin et al. (2001)](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=937655) paper.
### Citation Information
```bibtex
@inproceedings{martin2001database,
title={A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics},
author={Martin, David and Fowlkes, Charless and Tal, Doron and Malik, Jitendra},
booktitle={Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001},
volume={2},
pages={416--423},
year={2001},
organization={IEEE}
}
```
### Contributions
Thanks to [@eugenesiow](https://github.com/eugenesiow) for adding this dataset.
|
false |
# Dataset Card for Corpus Carolina
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** [sites.usp.br/corpuscarolina](https://sites.usp.br/corpuscarolina/)
- **Current Version:** 1.2 (Ada)
- **Point of Contact:** [LaViHD](mailto:lavihd@usp.br)
### Dataset Summary
Carolina is an Open Corpus for Linguistics and Artificial Intelligence with a
robust volume of texts of varied typology in contemporary Brazilian Portuguese
(1970-2021). This corpus contains documents and texts extracted from the web
and includes information (metadata) about its provenance and tipology.
The documents are clustered into taxonomies and the corpus can be loaded in complete or taxonomy modes. To load a single taxonomy, it is possible to pass a code as a parameter to the loading script (see the example bellow). Codes are 3-letters string and possible values are:
- `dat` : datasets and other corpora;
- `jud` : judicial branch;
- `leg` : legislative branch;
- `pub` : public domain works;
- `soc` : social media;
- `uni` : university domains;
- `wik` : wikis.
Dataset Vesioning:
The Carolina Corpus is under continuous development resulting in multiple vesions. The current version is v1.2, but v1.1 is also available. You can access diferent vesions of the corpus using the `revision` parameter on `load_dataset`.
Usage Example:
```python
from datasets import load_dataset
# to load all taxonomies
corpus_carolina = load_dataset("carolina-c4ai/corpus-carolina")
# to load social media documents
social_media = load_dataset("carolina-c4ai/corpus-carolina", taxonomy="soc")
# to load previous version
corpus_carolina = load_dataset("carolina-c4ai/corpus-carolina", revision="v1.1")
```
### Supported Tasks
Carolina corpus was compiled for academic purposes,
namely linguistic and computational analysis.
### Languages
Contemporary Brazilian Portuguese (1970-2021).
## Dataset Structure
Files are stored inside `corpus` folder with a subfolder
for each taxonomy. Every file folows a XML structure
(TEI P5) and contains multiple extracted documents. For
each document, the text and metadata are exposed as
`text` and `meta` features, respectively.
### Data Instances
Every instance have the following structure.
```
{
"meta": datasets.Value("string"),
"text": datasets.Value("string")
}
```
| Code | Taxonomy | Instances | Size |
|:----:|:---------------------------|----------:|-------:|
| | **Total** | 2107045 | 11 GB |
| dat | Datasets and other Corpora | 1102049 | 4.4 GB |
| wik | Wikis | 960139 | 5.2 GB |
| jud | Judicial Branch | 40464 | 1.5 GB |
| leg | Legislative Branch | 13 | 25 MB |
| soc | Social Media | 3413 | 17 MB |
| uni | University Domains | 941 | 10 MB |
| pub | Public Domain Works | 26 | 4.5 MB |
||
### Data Fields
- `meta`: a XML string with a TEI conformant `teiHeader` tag. It is exposed as text and needs to be parsed in order to access the actual metada;
- `text`: a string containing the extracted document.
### Data Splits
As a general corpus, Carolina does not have splits. In order to load the dataset, it is used `corpus` as its single split.
## Additional Information
### Dataset Curators
The Corpus Carolina is developed by a multidisciplinary
team of linguists and computer scientists, members of the
Virtual Laboratory of Digital Humanities - LaViHD and the Artificial Intelligence Center of the University of São Paulo - C4AI.
### Licensing Information
The Open Corpus for Linguistics and Artificial Intelligence (Carolina) was
compiled for academic purposes, namely linguistic and computational analysis.
It is composed of texts assembled in various digital repositories, whose
licenses are multiple and therefore should be observed when making use of the
corpus. The Carolina headers are licensed under Creative Commons
Attribution-NonCommercial-ShareAlike 4.0 International."
### Citation Information
```
@misc{corpusCarolinaV1.1,
title={
Carolina:
The Open Corpus for Linguistics and Artificial Intelligence
},
author={
Finger, Marcelo and
Paixão de Sousa, Maria Clara and
Namiuti, Cristiane and
Martins do Monte, Vanessa and
Costa, Aline Silva and
Serras, Felipe Ribas and
Sturzeneker, Mariana Lourenço and
Guets, Raquel de Paula and
Mesquita, Renata Morais and
Mello, Guilherme Lamartine de and
Crespo, Maria Clara Ramos Morales and
Rocha, Maria Lina de Souza Jeannine and
Brasil, Patrícia and
Silva, Mariana Marques da and
Palma, Mayara Feliciano
},
howpublished={\url{
https://sites.usp.br/corpuscarolina/corpus}},
year={2022},
note={Version 1.1 (Ada)},
}
```
|
false |
# Dataset Card for TupleInf Open IE
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Tuple IE Homepage](https://allenai.org/data/tuple-ie)
- **Repository:**
- **Paper:** [Answering Complex Questions Using Open Information Extraction](https://www.semanticscholar.org/paper/Answering-Complex-Questions-Using-Open-Information-Khot-Sabharwal/0ff595f0645a3e25a2f37145768985b10ead0509)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
The TupleInf Open IE dataset contains Open IE tuples extracted from 263K sentences that were used by the solver in “Answering Complex Questions Using Open Information Extraction” (referred as Tuple KB, T). These sentences were collected from a large Web corpus using training questions from 4th and 8th grade as queries. This dataset contains 156K sentences collected for 4th grade questions and 107K sentences for 8th grade questions. Each sentence is followed by the Open IE v4 tuples using their simple format.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
The text in the dataset is in English, collected from a large Web corpus using training questions from 4th and 8th grade as queries.
## Dataset Structure
### Data Instances
This dataset contains setences with corresponding relation tuples extracted from each sentence. Each instance should contain a sentence and followed by the [Open IE v4](https://github.com/allenai/openie-standalone) tuples using their *simple format*.
An example of an instance:
```JSON
{
"sentence": "0.04593 kg Used a triple beam balance to mass a golf ball.",
"tuples": {
"score": 0.8999999761581421,
"tuple_text": "(0.04593 kg; Used; a triple beam balance; to mass a golf ball)",
"context": "",
"arg1": "0.04593 kg",
"rel": "Used",
"arg2s": ["a triple beam balance", "to mass a golf ball"],
}
}
```
### Data Fields
- `sentence`: the input text/sentence.
- `tuples`: the extracted relation tuples from the sentence.
- `score`: the confident score for each tuple.
- `tuple_text`: the relationship representation text of the extraction, in the *simple format* of [Open IE v4](https://github.com/allenai/openie-standalone).
- `context`: an optional representation of the context for this extraction. Defaults to `""` if there's no context.
- `arg1`: the first argument in the relationship.
- `rel`: the relation.
- `arg2s`: a sequence of the 2nd arguments in the realtionship.
### Data Splits
| name | train|
|-----------|-----:|
| all |267719|
| 4th_grade |158910|
| 8th_grade |108809|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```bibtex
@article{Khot2017AnsweringCQ,
title={Answering Complex Questions Using Open Information Extraction},
author={Tushar Khot and A. Sabharwal and Peter Clark},
journal={ArXiv},
year={2017},
volume={abs/1704.05572}
}
```
### Contributions
Thanks to [@mattbui](https://github.com/mattbui) for adding this dataset. |
false |
# Dataset for training Russian language models
Overall: 75G
Scripts: https://github.com/IlyaGusev/rulm/tree/master/data_processing
| Website | Char count (M) | Word count (M) |
|-----------------|---------------|---------------|
| pikabu | 14938 | 2161 |
| lenta | 1008 | 135 |
| stihi | 2994 | 393 |
| stackoverflow | 1073 | 228 |
| habr | 5112 | 753 |
| taiga_fontanka | 419 | 55 |
| librusec | 10149 | 1573 |
| buriy | 2646 | 352 |
| ods_tass | 1908 | 255 |
| wiki | 3473 | 469 |
| math | 987 | 177 |
|
true |
# Dataset Card for 10k German News Articles Datasets
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [10k German News Article Dataset](https://tblock.github.io/10kGNAD/)
- **Repository:** [10k German News Article Dataset](https://github.com/tblock/10kGNAD)
- **Point of Contact:** [Steven Liu](stevhliu@gmail.com)
### Dataset Summary
The 10k German News Article Dataset consists of 10273 German language news articles from the online Austrian
newspaper website DER Standard. Each news article has been classified into one of 9 categories by professional
forum moderators employed by the newspaper. This dataset is extended from the original
[One Million Posts Corpus](https://ofai.github.io/million-post-corpus/). The dataset was created to support
topic classification in German because a classifier effective on a English dataset may not be as effective on
a German dataset due to higher inflections and longer compound words. Additionally, this dataset can be used
as a benchmark dataset for German topic classification.
### Supported Tasks and Leaderboards
This dataset can be used to train a model, like [BERT](https://huggingface.co/bert-base-uncased) for `topic classification` on German news articles. There are 9 possible categories.
### Languages
The text is in German and it comes from an online Austrian newspaper website. The BCP-47 code for German is
`de-DE`.
## Dataset Structure
### Data Instances
An example data instance contains a German news article (title and article are concatenated) and it's corresponding topic category.
```
{'text': ''Die Gewerkschaft GPA-djp lanciert den "All-in-Rechner" und findet, dass die Vertragsform auf die Führungsebene beschränkt gehört. Wien – Die Gewerkschaft GPA-djp sieht Handlungsbedarf bei sogenannten All-in-Verträgen.'
'label': 'Wirtschaft'
}
```
### Data Fields
* `text`: contains the title and content of the article
* `label`: can be one of 9 possible topic categories (`Web`, `Panorama`, `International`, `Wirtschaft`, `Sport`, `Inland`, `Etat`, `Wissenschaft`, `Kultur`)
### Data Splits
The data is split into a training set consisting of 9245 articles and a test set consisting of 1028 articles.
## Dataset Creation
### Curation Rationale
The dataset was created to support topic classification in the German language. English text classification datasets are common ([AG News](https://huggingface.co/datasets/ag_news) and [20 Newsgroup](https://huggingface.co/datasets/newsgroup)), but German datasets are less common. A classifier trained on an English dataset may not work as well on a set of German text due to grammatical differences. Thus there is a need for a German dataset for effectively assessing model performance.
### Source Data
#### Initial Data Collection and Normalization
The 10k German News Article Dataset is extended from the One Million Posts Corpus. 10273 German news articles were collected from this larger corpus. In the One Million Posts Corpus, each article has a topic path like
`Newsroom/Wirtschaft/Wirtschaftpolitik/Finanzmaerkte/Griechenlandkrise`. The 10kGNAD uses the second part of the topic path as the topic label. Article title and texts are concatenated into one text and author names are removed to avoid keyword classification on authors who write frequently on a particular topic.
#### Who are the source language producers?
The language producers are the authors of the Austrian newspaper website DER Standard.
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
This dataset was curated by Timo Block.
### Licensing Information
This dataset is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 license.
### Citation Information
Please consider citing the authors of the "One Million Post Corpus" if you use the dataset.:
```
@InProceedings{Schabus2017,
Author = {Dietmar Schabus and Marcin Skowron and Martin Trapp},
Title = {One Million Posts: A Data Set of German Online Discussions},
Booktitle = {Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)},
Pages = {1241--1244},
Year = {2017},
Address = {Tokyo, Japan},
Doi = {10.1145/3077136.3080711},
Month = aug
}
```
### Contributions
Thanks to [@stevhliu](https://github.com/stevhliu) for adding this dataset. |
false |
# Dataset Card for GEM/turku_paraphrase_corpus
## Dataset Description
- **Homepage:** https://turkunlp.org/paraphrase.html
- **Repository:** https://github.com/TurkuNLP/Turku-paraphrase-corpus
- **Paper:** https://aclanthology.org/2021.nodalida-main.29/
- **Leaderboard:** N/A
- **Point of Contact:** Jenna Kanerva, Filip Ginter
### Link to Main Data Card
You can find the main data card on the [GEM Website](https://gem-benchmark.com/data_cards/turku_paraphrase_corpus).
### Dataset Summary
This is a Finnish paraphrase corpus which consists of pairs of text passages, where a typical passage is about a sentence long. It can be used to either identify or generate paraphrases.
You can load the dataset via:
```
import datasets
data = datasets.load_dataset('GEM/turku_paraphrase_corpus')
```
The data loader can be found [here](https://huggingface.co/datasets/GEM/turku_paraphrase_corpus).
#### website
[Website](https://turkunlp.org/paraphrase.html)
#### paper
[ACL Anthology](https://aclanthology.org/2021.nodalida-main.29/)
#### authors
Jenna Kanerva, Filip Ginter, Li-Hsin Chang, Iiro Rastas, Valtteri Skantsi, Jemina Kilpeläinen, Hanna-Mari Kupari, Aurora Piirto, Jenna Saarni, Maija Sevón, Otto Tarkka (TurkuNLP / University of Turku)
## Dataset Overview
### Where to find the Data and its Documentation
#### Webpage
<!-- info: What is the webpage for the dataset (if it exists)? -->
<!-- scope: telescope -->
[Website](https://turkunlp.org/paraphrase.html)
#### Download
<!-- info: What is the link to where the original dataset is hosted? -->
<!-- scope: telescope -->
[Github](https://github.com/TurkuNLP/Turku-paraphrase-corpus)
#### Paper
<!-- info: What is the link to the paper describing the dataset (open access preferred)? -->
<!-- scope: telescope -->
[ACL Anthology](https://aclanthology.org/2021.nodalida-main.29/)
#### BibTex
<!-- info: Provide the BibTex-formatted reference for the dataset. Please use the correct published version (ACL anthology, etc.) instead of google scholar created Bibtex. -->
<!-- scope: microscope -->
```
@inproceedings{kanerva-etal-2021-finnish,
title = {Finnish Paraphrase Corpus},
author = {Kanerva, Jenna and Ginter, Filip and Chang, Li-Hsin and Rastas, Iiro and Skantsi, Valtteri and Kilpel{\"a}inen, Jemina and Kupari, Hanna-Mari and Saarni, Jenna and Sev{\'o}n, Maija and Tarkka, Otto},
booktitle = {Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa'21)},
year = {2021},
publisher = {Link{\"o}ping University Electronic Press, Sweden},
url = {https://aclanthology.org/2021.nodalida-main.29},
pages = {288--298}
}
```
#### Contact Name
<!-- quick -->
<!-- info: If known, provide the name of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
Jenna Kanerva, Filip Ginter
#### Contact Email
<!-- info: If known, provide the email of at least one person the reader can contact for questions about the dataset. -->
<!-- scope: periscope -->
jmnybl@utu.fi, figint@utu.fi
#### Has a Leaderboard?
<!-- info: Does the dataset have an active leaderboard? -->
<!-- scope: telescope -->
no
### Languages and Intended Use
#### Multilingual?
<!-- quick -->
<!-- info: Is the dataset multilingual? -->
<!-- scope: telescope -->
no
#### Covered Dialects
<!-- info: What dialects are covered? Are there multiple dialects per language? -->
<!-- scope: periscope -->
written standard language, spoken language
#### Covered Languages
<!-- quick -->
<!-- info: What languages/dialects are covered in the dataset? -->
<!-- scope: telescope -->
`Finnish`
#### License
<!-- quick -->
<!-- info: What is the license of the dataset? -->
<!-- scope: telescope -->
cc-by-sa-4.0: Creative Commons Attribution Share Alike 4.0 International
#### Intended Use
<!-- info: What is the intended use of the dataset? -->
<!-- scope: microscope -->
Paraphrase classification, paraphrase generation
#### Primary Task
<!-- info: What primary task does the dataset support? -->
<!-- scope: telescope -->
Paraphrasing
#### Communicative Goal
<!-- quick -->
<!-- info: Provide a short description of the communicative goal of a model trained for this task on this dataset. -->
<!-- scope: periscope -->
The corpus provides naturally occurring Finnish paraphrases striving for low lexical overlap, thus supporting many different downstream applications requiring language understanding.
### Credit
#### Curation Organization Type(s)
<!-- info: In what kind of organization did the dataset curation happen? -->
<!-- scope: telescope -->
`academic`
#### Curation Organization(s)
<!-- info: Name the organization(s). -->
<!-- scope: periscope -->
University of Turku
#### Dataset Creators
<!-- info: Who created the original dataset? List the people involved in collecting the dataset and their affiliation(s). -->
<!-- scope: microscope -->
Jenna Kanerva, Filip Ginter, Li-Hsin Chang, Iiro Rastas, Valtteri Skantsi, Jemina Kilpeläinen, Hanna-Mari Kupari, Aurora Piirto, Jenna Saarni, Maija Sevón, Otto Tarkka (TurkuNLP / University of Turku)
#### Funding
<!-- info: Who funded the data creation? -->
<!-- scope: microscope -->
The Turku paraphrase corpus project was funded by the Academy of Finland, as well as the European Language Grid project through its open call for pilot projects. The European Language Grid project has received funding from the European Union’s Horizon 2020 Research and Innovation programme under Grant Agreement no. 825627 (ELG).
#### Who added the Dataset to GEM?
<!-- info: Who contributed to the data card and adding the dataset to GEM? List the people+affiliations involved in creating this data card and who helped integrate this dataset into GEM. -->
<!-- scope: microscope -->
Jenna Kanerva, Filip Ginter (TurkuNLP / University of Turku)
### Dataset Structure
#### Data Fields
<!-- info: List and describe the fields present in the dataset. -->
<!-- scope: telescope -->
The dataset consist of pairs of text passages, where a typical passage is about a sentence long, however, a passage may also be longer or shorter than a sentence. Thus, each example include two text passages (string), a manually annotated label to indicate the paraphrase type (string), and additional metadata.
The dataset include three different `modes`, plain, classification, and generation. The `plain` mode loads the original data without any additional preprocessing or transformations, while the `classification` mode directly builds the data in a form suitable for training a paraphrase classifier, where each example is doubled in the data with different directions (text1, text2, label) --> (text2, text1, label) taking care of the label flipping as well if needed (paraphrases with directionality flag < or >). In the `generation` mode, the examples are preprocessed to be directly suitable for paraphrase generation task. In here, paraphrases not suitable for generation are discarded (negative, and highly context-dependent paraphrases), and directional paraphrases are provided so that the generation goes from more detailed passage to the more general one in order to prevent model hallucination (i.e. model learning to introduce new information). The rest of the paraphrases are provided in both directions (text1, text2, label) --> (text2, text1, label).
Each pair in `plain` and `classification` mode will include fields:
`gem_id`: Identifier of the paraphrase pair (string)
`goeswith`: Identifier of the document from which the paraphrase was extracted, can be `not available` in case the source of the paraphrase is not from document-structured data (string)
`fold`: 0-99, data split into 100 parts respecting document boundaries, you can use this e.g. to implement crossvalidation safely as all paraphrases from one document are in one fold (int)
`text1`: First paraphrase passage (string)
`text2`: Second paraphrase passage (string)
`label`: Manually annotated labels (string)
`binary_label`: Label turned into binary with values `positive` (paraphrase) and `negative` (not-paraphrase) (string)
`is_rewrite`: Indicator whether the example is human produced rewrite or naturally occurring paraphrase (bool)
Each pair in `generation` mode will include the same fields expect `text1` and `text2` are renamed to `input` and `output` in order to indicate the generation direction. Thus the fields are:
`gem_id`: Identifier of the paraphrase pair (string)
`goeswith`: Identifier of the document from which the paraphrase was extracted, can be `not available` in case the source of the paraphrase is not from document-structured data (string)
`fold`: 0-99, data split into 100 parts respecting document boundaries, you can use this e.g. to implement crossvalidation safely as all paraphrases from one document are in one fold (int)
`input`: The input paraphrase passage for generation (string)
`output`: The output paraphrase passage for generation (string)
`label`: Manually annotated labels (string)
`binary_label`: Label turned into binary with values `positive` (paraphrase) and `negative` (not-paraphrase) (string)
`is_rewrite`: Indicator whether the example is human produced rewrite or naturally occurring paraphrase (bool)
#### Example Instance
<!-- info: Provide a JSON formatted example of a typical instance in the dataset. -->
<!-- scope: periscope -->
```
{
'gem_id': 'gem-turku_paraphrase_corpus-train-15',
'goeswith': 'episode-02243',
'fold': 0,
'text1': 'Mitä merkitystä sillä on?',
'text2': 'Mitä väliä sillä edes on?',
'label': '4',
'binary_label': 'positive',
'is_rewrite': False
}
```
#### Data Splits
<!-- info: Describe and name the splits in the dataset if there are more than one. -->
<!-- scope: periscope -->
The corpus include 3 splits: train, validation, and test.
#### Splitting Criteria
<!-- info: Describe any criteria for splitting the data, if used. If there are differences between the splits (e.g., if the training annotations are machine-generated and the dev and test ones are created by humans, or if different numbers of annotators contributed to each example), describe them here. -->
<!-- scope: microscope -->
The data is split randomly into the three section with a restriction of all paraphrases from the same document (movie, TV episode, news article, student translation, or exam question) being in the same section. All splits are manually annotated.
## Dataset in GEM
### Rationale for Inclusion in GEM
#### Why is the Dataset in GEM?
<!-- info: What does this dataset contribute toward better generation evaluation and why is it part of GEM? -->
<!-- scope: microscope -->
This dataset provides a large amount of high quality (manually collected and verified) paraphrases for Finnish.
#### Similar Datasets
<!-- info: Do other datasets for the high level task exist? -->
<!-- scope: telescope -->
yes
#### Unique Language Coverage
<!-- info: Does this dataset cover other languages than other datasets for the same task? -->
<!-- scope: periscope -->
no
#### Ability that the Dataset measures
<!-- info: What aspect of model ability can be measured with this dataset? -->
<!-- scope: periscope -->
natural language understanding, language variation
### GEM-Specific Curation
#### Modificatied for GEM?
<!-- info: Has the GEM version of the dataset been modified in any way (data, processing, splits) from the original curated data? -->
<!-- scope: telescope -->
yes
#### GEM Modifications
<!-- info: What changes have been made to he original dataset? -->
<!-- scope: periscope -->
`data points modified`
#### Modification Details
<!-- info: For each of these changes, described them in more details and provided the intended purpose of the modification -->
<!-- scope: microscope -->
Data structure is slightly simplified, and the release provides ready made transformations into two tasks (paraphrase classification and generation), where some data instances are doubled with different direction, and some are discarded as not being suitable for generation (e.g. negatives).
#### Additional Splits?
<!-- info: Does GEM provide additional splits to the dataset? -->
<!-- scope: telescope -->
no
### Getting Started with the Task
## Previous Results
### Previous Results
#### Measured Model Abilities
<!-- info: What aspect of model ability can be measured with this dataset? -->
<!-- scope: telescope -->
natural language understanding, language variation
#### Previous results available?
<!-- info: Are previous results available? -->
<!-- scope: telescope -->
yes
#### Other Evaluation Approaches
<!-- info: What evaluation approaches have others used? -->
<!-- scope: periscope -->
F-score in paraphrase classification
## Dataset Curation
### Original Curation
#### Original Curation Rationale
<!-- info: Original curation rationale -->
<!-- scope: telescope -->
The dataset is fully manually annotated. The dataset strives for interesting paraphrases with low lexical overlap, thus the annotation is two fold. First the paraphrases are manually extracted from two related documents, where the annotators are instructed to extract only interesting paraphrases. In the second phrase, all extracted paraphrases are manually labeled given the annotation scheme.
The annotation scheme is:
4 : paraphrase in all reasonably possible contexts
3 : paraphrase in the given document contexts, but not in general
2 : related but not paraphrase
During annotation also labels 1 (unrelated) and x (skip, e.g. wrong language) were used, however, the insignificant amount of examples annotated with these labels were discarded from the released corpus.
The following flags are annotated to label 4 paraphrases:
< : txt1 is more general than txt2; txt2 is more specific than txt1 (directional paraphrase where txt2 can be replaced with txt1 in all contexts but not to the other direction)
> : txt2 is more general than txt1; txt1 is more specific than txt2 (directional paraphrase where txt1 can be replaced with txt2 in all contexts but not to the other direction)
i : minor traceable difference (differing in terms of grammatical number or case, 'this' vs 'that', etc.)
s : style or strength difference (e.g. equivalent meaning, but one of the statements substantially more colloquial than the other)
For paraphrases where the annotated label was something else than label 4 without any flags, the annotators had an option to rewrite the text passages so that the rewritten paraphrase pair formed label 4 (universal) paraphrase. This was used for cases where simple edit would turn e.g. contextual or directional paraphrase into universal one. For the rewritten examples, both the original and the rewritten pairs are available with corresponding labels annotated.
#### Communicative Goal
<!-- info: What was the communicative goal? -->
<!-- scope: periscope -->
Representing text passages with identical meaning but different surface realization.
#### Sourced from Different Sources
<!-- info: Is the dataset aggregated from different data sources? -->
<!-- scope: telescope -->
yes
#### Source Details
<!-- info: List the sources (one per line) -->
<!-- scope: periscope -->
movie and TV series subtitles (82%)
news articles (9%)
discussion forum messages (8%)
university translation exercises (1%)
university course essays and exams (<1%)
### Language Data
#### How was Language Data Obtained?
<!-- info: How was the language data obtained? -->
<!-- scope: telescope -->
`Found`, `Other`
#### Where was it found?
<!-- info: If found, where from? -->
<!-- scope: telescope -->
`Multiple websites`, `Offline media collection`, `Other`
#### Language Producers
<!-- info: What further information do we have on the language producers? -->
<!-- scope: microscope -->
The movie and TV series subtitles are extracted from OPUS OpenSubtitles2018 collection, which is based on data from [OpenSubtitles](http://www.opensubtitles.org/).
The news articles are collected from two Finnish news sites, YLE and HS, during years 2017-2020.
Discussion forum messages are obtained from the Finnish Suomi24 discussion forum released for academic use (http://urn.fi/urn:nbn:fi:lb-2020021801).
University translation exercises, essays and exams are collected during the project.
#### Data Validation
<!-- info: Was the text validated by a different worker or a data curator? -->
<!-- scope: telescope -->
validated by data curator
#### Was Data Filtered?
<!-- info: Were text instances selected or filtered? -->
<!-- scope: telescope -->
not filtered
### Structured Annotations
#### Additional Annotations?
<!-- quick -->
<!-- info: Does the dataset have additional annotations for each instance? -->
<!-- scope: telescope -->
expert created
#### Number of Raters
<!-- info: What is the number of raters -->
<!-- scope: telescope -->
2<n<10
#### Rater Qualifications
<!-- info: Describe the qualifications required of an annotator. -->
<!-- scope: periscope -->
Members of the TurkuNLP research group, native speakers of Finnish, each annotator has a strong background in language studies by having an academic degree or ongoing studies in a field related to languages or linguistics.
#### Raters per Training Example
<!-- info: How many annotators saw each training example? -->
<!-- scope: periscope -->
1
#### Raters per Test Example
<!-- info: How many annotators saw each test example? -->
<!-- scope: periscope -->
1
#### Annotation Service?
<!-- info: Was an annotation service used? -->
<!-- scope: telescope -->
no
#### Annotation Values
<!-- info: Purpose and values for each annotation -->
<!-- scope: microscope -->
1. Manual extraction of interesting paraphrases from two related documents.
2. Manual labeling of each extracted paraphrase based on the given annotation scheme, e.g. distinguishing contextual and universal paraphrases, marking style or strength differences, etc.
#### Any Quality Control?
<!-- info: Quality control measures? -->
<!-- scope: telescope -->
validated by another rater
#### Quality Control Details
<!-- info: Describe the quality control measures that were taken. -->
<!-- scope: microscope -->
Partial double annotation, double annotation batches are assigned regularly in order to monitor annotation consistency. In double annotation, one annotator first extracts the candidate paraphrases, and these candidates are assigned to two different annotators, who does the label annotation independently from each other. Afterwards, the label annotations are merged, and conflicting labels are resolved together with the whole annotation team.
### Consent
#### Any Consent Policy?
<!-- info: Was there a consent policy involved when gathering the data? -->
<!-- scope: telescope -->
yes
#### Consent Policy Details
<!-- info: What was the consent policy? -->
<!-- scope: microscope -->
The corpus is mostly based on public/open data. For other data sources (student material), the licensing was agreed with the data providers during the collection.
### Private Identifying Information (PII)
#### Contains PII?
<!-- quick -->
<!-- info: Does the source language data likely contain Personal Identifying Information about the data creators or subjects? -->
<!-- scope: telescope -->
likely
#### Categories of PII
<!-- info: What categories of PII are present or suspected in the data? -->
<!-- scope: periscope -->
`generic PII`
#### Any PII Identification?
<!-- info: Did the curators use any automatic/manual method to identify PII in the dataset? -->
<!-- scope: periscope -->
no identification
### Maintenance
#### Any Maintenance Plan?
<!-- info: Does the original dataset have a maintenance plan? -->
<!-- scope: telescope -->
no
## Broader Social Context
### Previous Work on the Social Impact of the Dataset
#### Usage of Models based on the Data
<!-- info: Are you aware of cases where models trained on the task featured in this dataset ore related tasks have been used in automated systems? -->
<!-- scope: telescope -->
no
### Impact on Under-Served Communities
#### Addresses needs of underserved Communities?
<!-- info: Does this dataset address the needs of communities that are traditionally underserved in language technology, and particularly language generation technology? Communities may be underserved for exemple because their language, language variety, or social or geographical context is underepresented in NLP and NLG resources (datasets and models). -->
<!-- scope: telescope -->
no
### Discussion of Biases
#### Any Documented Social Biases?
<!-- info: Are there documented social biases in the dataset? Biases in this context are variations in the ways members of different social categories are represented that can have harmful downstream consequences for members of the more disadvantaged group. -->
<!-- scope: telescope -->
no
## Considerations for Using the Data
### PII Risks and Liability
#### Potential PII Risk
<!-- info: Considering your answers to the PII part of the Data Curation Section, describe any potential privacy to the data subjects and creators risks when using the dataset. -->
<!-- scope: microscope -->
None
### Licenses
#### Copyright Restrictions on the Dataset
<!-- info: Based on your answers in the Intended Use part of the Data Overview Section, which of the following best describe the copyright and licensing status of the dataset? -->
<!-- scope: periscope -->
`open license - commercial use allowed`
#### Copyright Restrictions on the Language Data
<!-- info: Based on your answers in the Language part of the Data Curation Section, which of the following best describe the copyright and licensing status of the underlying language data? -->
<!-- scope: periscope -->
`open license - commercial use allowed`
### Known Technical Limitations
|
false | # MixInstruct
## Introduction
This is the official realease of dataset **MixInstruct** for project **LLM-Blender**.
This dataset contains 11 responses from the current popular instruction following-LLMs that includes:
1. [Stanford Alpaca](https://huggingface.co/chavinlo/alpaca-native)
2. [FastChat Vicuna](https://huggingface.co/eachadea/vicuna-13b-1.1)
3. [Dolly V2](https://huggingface.co/databricks/dolly-v2-12b)
4. [StableLM](https://huggingface.co/stabilityai/stablelm-tuned-alpha-7b)
5. [Open Assistant](https://huggingface.co/OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5)
6. [Koala](https://huggingface.co/TheBloke/koala-7B-HF)
7. [Baize](https://huggingface.co/mosesjun0h/llama-7b-hf-baize-lora-bf16)
8. [Flan-T5](https://huggingface.co/google/flan-t5-xxl)
9. [ChatGLM](https://huggingface.co/THUDM/chatglm-6b)
10. [MOSS](https://huggingface.co/fnlp/moss-moon-003-sft)
11. [Moasic MPT](https://huggingface.co/mosaicml/mpt-7b-instruct)
We evaluate each response with auto metrics including BLEU, ROUGE, BERTScore, BARTScore. And provide pairwise comparison results by prompting ChatGPT for the $4771$ examples
in the test split. (The rest $229$ examples contain contents filtered by the API).
## Data Format
```json
[
{
"id": "unified_chip2/69962",
"instruction": "",
"input": "I've always wondered what the difference is between a skeptic and a denier.",
"output": "A skeptic is someone who questions the validity of something, while a denier is someone who outright rejects something without evidence or reason.",
"candidates": [
{
"decoding_method": "top_p_sampling",
"model": "oasst-sft-4-pythia-12b-epoch-3.5",
"text": "A skeptic is someone who doubts or expresses ...",
"scores": {
"logprobs": -0.02404022216796875,
"bleu": 5.656152750894142,
"bertscore": 0.7549101114273071,
"rouge1": 0.2857142857142857,
"rouge2": 0.1272727272727273,
"rougeL": 0.23214285714285715,
"rougeLsum": 0.23214285714285715
}
},
...
],
},
...
]
```
Examples evaluted by ChatGPT will contain another filed **cmp_results**.
The options contains:
1. A is better
2. B is better
3. Same good
4. Same bad
```json
"cmp_results": {
"model_A,model_B": "A is better",
...
},
```
Each cmp_results field is encoded into a str in a json format. Please first use `json.loads(item['cmp_results'])` to get the cmp_results for each item.
"null" denotes no cmp_results from ChatGPT avaliable.
## Eval Results
### Auto Metrics
- train
| Models (down) / Metircs (right) | logprobs | rougeL | rouge2 | rougeLsum | rouge1 | bleu | bertscore | bleurt | bartscore |
|:----------------------------------|:------------|:----------------|:----------------|:----------------|:----------------|:----------------|:----------------|:----------------|:-------------|
| alpaca-native | -6.1247 | 0.248 | 0.1414 | 0.2986 | 0.3347 | 8.057 | 0.7196 | -0.5092 | -3.5335 |
| chatglm-6b | -10.1263 | 0.2231 | 0.1212 | 0.2743 | 0.3074 | 6.2597 | 0.7043 | -0.6071 | -3.4975 |
| dolly-v2-12b | -24.8508 | 0.1245 | 0.0502 | 0.1625 | 0.1836 | 2.1062 | 0.6244 | -0.8562 | -3.8145 |
| flan-t5-xxl | -1.0717 | 0.1202 | 0.0456 | 0.1334 | 0.1489 | 1.8418 | 0.6514 | -1.2176 | -4.537 |
| koala-7B-HF | -10.8323 | 0.1533 | 0.0683 | 0.1909 | 0.2165 | 3.2848 | 0.6436 | -0.8284 | -3.8326 |
| llama-7b-hf-baize-lora-bf16 | -24.8867 | 0.1539 | 0.0797 | 0.2042 | 0.2276 | 3.4928 | 0.6564 | -0.6575 | -3.496 |
| moss-moon-003-sft | -796.1366 | 0.1599 | 0.0898 | 0.2135 | 0.236 | 3.944 | 0.6689 | -0.5617 | -3.3404 |
| mpt-7b | -174.1702 | 0.1118 | 0.0447 | 0.1517 | 0.1683 | 1.7698 | 0.618 | -0.9525 | -3.9119 |
| mpt-7b-instruct | -156.8005 | 0.1225 | 0.0538 | 0.1669 | 0.1861 | 2.1041 | 0.6327 | -0.8176 | -3.6996 |
| oasst-sft-4-pythia-12b-epoch-3.5 | -4.7714 | 0.2902 | 0.1763 | 0.3447 | 0.386 | 10.6599 | 0.748 | -0.3762 | -3.4221 |
| stablelm-tuned-alpha-7b | -1268.9396 | 0.1336 | 0.0544 | 0.1714 | 0.1948 | 2.6348 | 0.6355 | -0.9585 | -4.0795 |
| vicuna-13b-1.1 | -11.1528 | 0.211 | 0.1219 | 0.2671 | 0.3003 | 6.3697 | 0.6928 | -0.6194 | -3.4233 |
| Best Model Metric Perf | -1.0717 | 0.2902 | 0.1763 | 0.3447 | 0.386 | 10.6599 | 0.748 | -0.3762 | -3.3404 |
| Oracle | 0.0 | 0.3611 | 0.2471 | 0.4242 | 0.4706 | 15.8557 | 0.7783 | 0.0723 | 0.0 |
| Oracle-Best_Model Gap | 1.0717 | 0.0709 | 0.0708 | 0.0794 | 0.0846 | 5.1958 | 0.0303 | 0.4484 | 3.3404 |
- val
| Models (down) / Metircs (right) | logprobs | rouge1 | rouge2 | rougeLsum | rougeL | bleu | bertscore | bleurt | bartscore |
|:----------------------------------|:------------|:----------------|:----------------|:----------------|:----------------|:----------------|:----------------|:----------------|:---------------|
| alpaca-native | -3.3832 | 0.3342 | 0.1452 | 0.299 | 0.2503 | 8.1749 | 0.7198 | -0.5076 | -3.5517 |
| chatglm-6b | -4.7033 | 0.3066 | 0.1216 | 0.2743 | 0.2241 | 6.3323 | 0.7053 | -0.6091 | -3.51 |
| dolly-v2-12b | -9.1237 | 0.1843 | 0.0511 | 0.1633 | 0.1254 | 2.1368 | 0.6257 | -0.852 | -3.8121 |
| flan-t5-xxl | -1.0077 | 0.1497 | 0.0464 | 0.1342 | 0.1212 | 1.8653 | 0.652 | -1.2089 | -4.5407 |
| koala-7B-HF | -6.015 | 0.2154 | 0.068 | 0.1903 | 0.1538 | 3.2596 | 0.6425 | -0.8298 | -3.8456 |
| llama-7b-hf-baize-lora-bf16 | -12.2594 | 0.2261 | 0.0803 | 0.2034 | 0.1543 | 3.5462 | 0.6562 | -0.6604 | -3.4831 |
| moss-moon-003-sft | -357.3054 | 0.2053 | 0.0678 | 0.1851 | 0.1361 | 2.9639 | 0.648 | -0.7261 | -3.6317 |
| mpt-7b | -171.9416 | 0.1663 | 0.0447 | 0.1499 | 0.1111 | 1.7555 | 0.617 | -0.964 | -3.9189 |
| mpt-7b-instruct | -157.1143 | 0.1841 | 0.054 | 0.1652 | 0.1224 | 2.1252 | 0.6307 | -0.8275 | -3.7183 |
| oasst-ft-4-pythia-12b-epoch-3.5 | -1.6194 | 0.3835 | 0.1761 | 0.3434 | 0.2896 | 10.5858 | 0.7479 | -0.378 | -3.4366 |
| stablelm-tuned-alpha-7b | -869.6767 | 0.192 | 0.0529 | 0.1688 | 0.1317 | 2.5687 | 0.6314 | -0.9618 | -4.1008 |
| vicuna-13b-1.1 | -5.6143 | 0.3029 | 0.1242 | 0.2701 | 0.2142 | 6.5299 | 0.695 | -0.6212 | -3.4332 |
| Best Model Metric Perf | -1.0077 | 0.3835 | 0.1761 | 0.3434 | 0.2896 | 10.5858 | 0.7479 | -0.378 | -3.4332 |
| Oracle | 0.0 | 0.4712 | 0.2488 | 0.4258 | 0.3642 | 15.9896 | 0.7794 | 0.0726 | 0.0 |
| Oracle-Best_Model Gap | 1.0077 | 0.0877 | 0.0728 | 0.0824 | 0.0746 | 5.4038 | 0.0315 | 0.4506 | 3.4332 |
- test
| Models (down) / Metircs (right) | logprobs | rougeL | rougeLsum | rouge1 | rouge2 | bleu | bertscore | bleurt | bartscore |
|:----------------------------------|:------------|:----------------|:----------------|:----------------|:----------------|:----------------|:----------------|:----------------|:---------------|
| alpaca-native | -3.458 | 0.2421 | 0.2915 | 0.3276 | 0.1362 | 7.6478 | 0.7146 | -0.5307 | -3.5696 |
| chatglm-6b | -4.7418 | 0.2225 | 0.2734 | 0.3063 | 0.1192 | 6.0493 | 0.7038 | -0.6167 | -3.5193 |
| dolly-v2-12b | -9.1266 | 0.1236 | 0.1606 | 0.1811 | 0.0495 | 2.062 | 0.6226 | -0.8654 | -3.8331 |
| flan-t5-xxl | -0.9924 | 0.1172 | 0.1296 | 0.1444 | 0.0432 | 1.6066 | 0.6492 | -1.2288 | -4.5717 |
| koala-7B-HF | -6.1159 | 0.1507 | 0.1871 | 0.2131 | 0.0662 | 3.0983 | 0.6396 | -0.8354 | -3.8496 |
| llama-7b-hf-baize-lora-bf16 | -11.9519 | 0.1521 | 0.2022 | 0.2253 | 0.0781 | 3.4005 | 0.6557 | -0.663 | -3.526 |
| moss-moon-003-sft | -356.8774 | 0.1365 | 0.1863 | 0.2062 | 0.0686 | 2.9561 | 0.6485 | -0.7261 | -3.6461 |
| mpt-7b | -176.2144 | 0.1106 | 0.1498 | 0.1663 | 0.0439 | 1.7392 | 0.6165 | -0.9636 | -3.9419 |
| mpt-7b-instruct | -156.0153 | 0.121 | 0.1647 | 0.1837 | 0.0524 | 2.0692 | 0.6321 | -0.8232 | -3.7208 |
| oasst-sft-4-pythia-12b-epoch-3.5 | -1.6749 | 0.2873 | 0.341 | 0.3813 | 0.1738 | 10.5046 | 0.7468 | -0.3908 | -3.4486 |
| stablelm-tuned-alpha-7b | -831.595 | 0.1306 | 0.1672 | 0.1904 | 0.0524 | 2.5044 | 0.6247 | -0.9832 | -4.1208 |
| vicuna-13b-1.1 | -5.6914 | 0.2122 | 0.2677 | 0.3012 | 0.1223 | 6.3584 | 0.696 | -0.6146 | -3.4368 |
| Best Model Metric Perf | -0.9924 | 0.2873 | 0.341 | 0.3813 | 0.1738 | 10.5046 | 0.7468 | -0.3908 | -3.4368 |
| Oracle | 0.0 | 0.3585 | 0.4201 | 0.466 | 0.2438 | 15.4971 | 0.7767 | 0.0679 | 0.0 |
| Oracle-Best_Model Gap | 0.9924 | 0.0712 | 0.0791 | 0.0847 | 0.07 | 4.9925 | 0.0299 | 0.4587 | 3.4368 |
### ChatGPT CMPTS (4771 examples)
| **Methods** | BERTScore | BARTScore | BLEURT | GPT-Rank | Beat Vic(%) | Beat OA(%) | Top-1(%) | Top-2(%) | Top-3(%) |
|:-----------------:|:---------:|:---------:|:---------:|:--------:|:----------:|:----------:|:----------:|:----------:|:----------:|
| Open Assistant | **74.68** | -3.45 | **-0.39** | **3.90** | **62.78** | N/A | 17.35 | 35.67 | 51.98 |
| Vicuna | 69.60 | **-3.44** | -0.61 | 4.13 | N/A | **64.77** | **25.47** | **41.23** | **52.88** |
| Alpaca | 71.46 | -3.57 | -0.53 | 4.62 | 56.70 | 61.35 | 15.41 | 29.81 | 44.46 |
| Baize | 65.57 | -3.53 | -0.66 | 4.86 | 52.76 | 56.40 | 14.23 | 26.91 | 38.80 |
| moss | 64.85 | -3.65 | -0.73 | 5.09 | 51.62 | 51.79 | 15.93 | 27.52 | 38.27 |
| ChatGLM | 70.38 | -3.52 | -0.62 | 5.63 | 44.04 | 45.67 | 9.41 | 19.37 | 28.78 |
| Koala | 63.96 | -3.85 | -0.84 | 6.76 | 39.93 | 39.01 | 8.15 | 15.72 | 22.55 |
| Dolly v2 | 62.26 | -3.83 | -0.87 | 6.90 | 33.33 | 31.44 | 5.16 | 10.06 | 16.45 |
| Mosaic MPT | 63.21 | -3.72 | -0.82 | 7.19 | 30.87 | 30.16 | 5.39 | 10.61 | 16.24 |
| StableLM | 62.47 | -4.12 | -0.98 | 8.71 | 21.55 | 19.87 | 2.33 | 4.74 | 7.96 |
| Flan-T5 | 64.92 | -4.57 | -1.23 | 8.81 | 23.89 | 19.93 | 1.30 | 2.87 | 5.32 |
| Oracle(BERTScore) | **77.67** | -3.17 | -0.27 | 3.88 | 54.41 | 38.84 | 20.16 | 38.11 | 53.49 |
| Oracle(BLEURT) | 75.02 | -3.15 | **-0.15** | 3.77 | 55.61 | 45.80 | 21.48 | 39.84 | 55.36 |
| Oracle(BARTScore) | 73.23 | **-2.87** | -0.38 | 3.69 | 50.32 | 57.01 | 26.10 | 43.70 | 57.33 |
| Oracle(ChatGPT) | 70.32 | -3.33 | -0.51 | **1.00** | **100.00** | **100.00** | **100.00** | **100.00** | **100.00** |
|
false |
# Dataset Card for mC4-es-sampled
## Table of Contents
- [Dataset Card for mC4-es-sampled](#dataset-card-for-mc4-es-sampled)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://huggingface.co/datasets/allenai/c4
- **Paper:** https://arxiv.org/abs/1910.10683
### Dataset Summary
This dataset is the result of applying perplexity sampling to the Spanish portion of mC4 using [`mc4-sampling`](https://huggingface.co/datasets/bertin-project/mc4-sampling/). Please, refer to [BERTIN Project](https://huggingface.co/bertin-project/bertin-roberta-base-spanish).
You can load the mC4 Spanish sampled like this:
```python
from datasets import load_dataset
for config in ("random", "stepwise", "gaussian"):
mc4es = load_dataset(
"bertin-project/mc4-es-sampled",
config,
split="train",
streaming=True
).shuffle(buffer_size=1000)
for sample in mc4es:
print(config, sample)
break
```
Alternatively, you can bypass the `datasets` library and quickly download (\~1.5hrs, depending on connection) a specific config in the same order used to pre-train BERTIN models in a massive (\~200GB) JSON-lines files:
```python
import io
import gzip
import json
import sys
import requests
from tqdm import tqdm
_DATA_URL_TRAIN = "https://huggingface.co/datasets/bertin-project/mc4-es-sampled/resolve/main/mc4-es-train-50M-{config}-shard-{index:04d}-of-{n_shards:04d}.json.gz"
def main(config="stepwise"):
data_urls = [
_DATA_URL_TRAIN.format(
config=config,
index=index + 1,
n_shards=1024,
)
for index in range(1024)
]
with open(f"mc4-es-train-50M-{config}.jsonl", "w") as f:
for dara_url in tqdm(data_urls):
response = requests.get(dara_url)
bio = io.BytesIO(response.content)
with gzip.open(bio, "rt", encoding="utf8") as g:
for line in g:
json_line = json.loads(line.strip())
f.write(json.dumps(json_line) + "\
")
if __name__ == "__main__":
main(sys.argv[1])
```
### Supported Tasks and Leaderboards
mC4-es-sampled is mainly intended for reproducibility purposes of the BERTIN Project and to pretrain language models and word representations on medium budgets.
### Languages
The dataset only supports the Spanish language.
## Dataset Structure
### Data Instances
An example form the `Gaussian` config:
```python
{'timestamp': '2018-10-20T06:20:53Z', 'text': 'Ortho HyaluroTop 200 aporta el colágeno y ácido hialurónico que, con la edad, se producen en menor cantidad. La vitamina C promueve la producción de colágeno para mantener la piel sana y protege a las células contra los radicales libres causados ??por la contaminación ambiental y los rayos UV.', 'url': 'https://www.farmaciagaleno.com/orthonat-hyalurotop-200-30-capsulas'}
```
### Data Fields
The data have several fields:
- `url`: url of the source as a string
- `text`: text content as a string
- `timestamp`: timestamp as a string
### Data Splits
The resulting mC4 subsets for Spanish are reported in this table:
| config | train |
|:---------|:--------|
| stepwise | 50M |
| random | 50M |
| gaussian | 50M |
The split `validation` is exactly the same as the original `mc4` dataset.
## Dataset Creation
### Curation Rationale
This dataset was built from the original [`mc4`](https://huggingface.co/datasets/mc4) by applying perplexity-sampling via [`mc4-sampling`](https://huggingface.co/datasets/bertin-project/mc4-sampling) for Spanish.
## Additional Information
### Dataset Curators
Original data by [Common Crawl](https://commoncrawl.org/).
### Licensing Information
AllenAI are releasing this dataset under the terms of ODC-BY. By using this, you are also bound by the Common Crawl terms of use in respect of the content contained in the dataset.
### Citation Information
To cite this dataset ([arXiv](https://arxiv.org/abs/2207.06814)):
```bibtex
@article{BERTIN,
author = {Javier De la Rosa y Eduardo G. Ponferrada y Manu Romero y Paulo Villegas y Pablo González de Prado Salas y María Grandury},
title = {{BERTIN}: Efficient Pre-Training of a Spanish Language Model using Perplexity Sampling},
journal = {Procesamiento del Lenguaje Natural},
volume = {68},
number = {0},
year = {2022},
keywords = {},
abstract = {The pre-training of large language models usually requires massive amounts of resources, both in terms of computation and data. Frequently used web sources such as Common Crawl might contain enough noise to make this pretraining sub-optimal. In this work, we experiment with different sampling methods from the Spanish version of mC4, and present a novel data-centric technique which we name perplexity sampling that enables the pre-training of language models in roughly half the amount of steps and using one fifth of the data. The resulting models are comparable to the current state-of-the-art, and even achieve better results for certain tasks. Our work is proof of the versatility of Transformers, and paves the way for small teams to train their models on a limited budget.},
issn = {1989-7553},
url = {http://journal.sepln.org/sepln/ojs/ojs/index.php/pln/article/view/6403},
pages = {13--23}
}
```
If you use this dataset, we would love to hear about it! Reach out on twitter, GitHub, Discord, or shoot us an email.
To cite the original `mc4` dataset:
```
@article{2019t5,
author = {Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu},
title = {Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer},
journal = {arXiv e-prints},
year = {2019},
archivePrefix = {arXiv},
eprint = {1910.10683},
}
```
### Contributions
Dataset contributed by [@versae](https://github.com/versae) for BERTIN Project.
Thanks to [@dirkgr](https://github.com/dirkgr) and [@lhoestq](https://github.com/lhoestq) for adding the original mC4 dataset.
|
true |
http://decomp.io/ |
false |
# Dataset Card for Bilingual Corpus of Arabic-English Parallel Tweets
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Bilingual Corpus of Arabic-English Parallel Tweets](https://alt.qcri.org/resources/bilingual_corpus_of_parallel_tweets)
- **Repository:**
- **Paper:** [Aclweb](https://www.aclweb.org/anthology/2020.bucc-1.3/)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
Twitter users often post parallel tweets—tweets that contain the same content but are written in different languages. Parallel tweets can be an important resource for developing machine translation (MT) systems among other natural language processing (NLP) tasks. This resource is a result of a generic method for collecting parallel tweets. Using the method, we compiled a bilingual corpus of English-Arabic parallel tweets and a list of Twitter accounts who post English-Arabic tweets regularly. Additionally, we annotate a subset of Twitter accounts with their countries of origin and topic of interest, which provides insights about the population who post parallel tweets.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
parallelTweets:
```
{
"ArabicTweetID": 981111245209243600,
"EnglishTweetID": 981111450432401400
}
```
accountList:
```
{
'account': 'HukoomiQatar'
}
```
countryTopicAnnotation:
```
{
'account': 'HukoomiQatar',
'country': 'QA',
'topic': 'Gov'
}
```
### Data Fields
parallelTweets:
- `ArabicTweetID` (int)
- `EnglishTweetID` (int)
accountList:
- `account` (str)
countryTopicAnnotation:
- `account` (str)
- `country` (class label): One of:
- "QA",
- "BH",
- "AE",
- "OM",
- "SA",
- "PL",
- "JO",
- "IQ",
- "Other",
- "EG",
- "KW",
- "SY"
- `topic` (class label): One of:
- "Gov",
- "Culture",
- "Education",
- "Sports",
- "Travel",
- "Events",
- "Business",
- "Science",
- "Politics",
- "Health",
- "Governoment",
- "Media",
### Data Splits
All configuration have only one split: "test".
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
It is licensed under the [Apache License, Version 2.0](http://www.apache.org/licenses/LICENSE-2.0).
### Citation Information
```
@inproceedings{Mubarak2020bilingualtweets,
title={Constructing a Bilingual Corpus of Parallel Tweets},
author={Mubarak, Hamdy and Hassan, Sabit and Abdelali, Ahmed},
booktitle={Proceedings of 13th Workshop on Building and Using Comparable Corpora (BUCC)},
address={Marseille, France},
year={2020}
}
```
[More Information Needed]
### Contributions
Thanks to [@sumanthd17](https://github.com/sumanthd17) for adding this dataset. |
false |
# Dataset Card for SciELO
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**[SciELO](https://sites.google.com/view/felipe-soares/datasets#h.p_92uSCyAjWSRB)
- **Repository:**
- **Paper:** [A Large Parallel Corpus of Full-Text Scientific Articles](https://arxiv.org/abs/1905.01852)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
A parallel corpus of full-text scientific articles collected from Scielo database in the following languages:English, Portuguese and Spanish.
The corpus is sentence aligned for all language pairs, as well as trilingual aligned for a small subset of sentences.
Alignment was carried out using the Hunalign algorithm.
### Supported Tasks and Leaderboards
The underlying task is machine translation.
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{soares2018large,
title={A Large Parallel Corpus of Full-Text Scientific Articles},
author={Soares, Felipe and Moreira, Viviane and Becker, Karin},
booktitle={Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC-2018)},
year={2018}
}
```
### Contributions
Thanks to [@patil-suraj](https://github.com/patil-suraj) for adding this dataset. |
false |
# Dataset Card for SO StackSample
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://www.kaggle.com/stackoverflow/stacksample
### Dataset Summary
Dataset with the text of 10% of questions and answers from the Stack Overflow programming Q&A website.
This is organized as three tables:
Questions table contains the title, body, creation date, closed date (if applicable), score, and owner ID for all non-deleted Stack Overflow questions whose Id is a multiple of 10.
Answers table contains the body, creation date, score, and owner ID for each of the answers to these questions. The ParentId column links back to the Questions table.
Tags table contains the tags on each of these questions.
### Supported Tasks and Leaderboards
Example projects include:
- Identifying tags from question text
- Predicting whether questions will be upvoted, downvoted, or closed based on their text
- Predicting how long questions will take to answer
- Open Domain Q/A
### Languages
English (en) and Programming Languages.
## Dataset Structure
### Data Instances
For Answers:
```
{
"Id": { # Unique ID given to the Answer post
"feature_type": "Value",
"dtype": "int32"
},
"OwnerUserId": { # The UserID of the person who generated the Answer on StackOverflow. -1 means NA
"feature_type": "Value",
"dtype": "int32"
},
"CreationDate": { # The date the Answer was generated. Follows standard datetime format.
"feature_type": "Value",
"dtype": "string"
},
"ParentId": { # Refers to the `Id` of the Question the Answer belong to.
"feature_type": "Value",
"dtype": "int32"
},
"Score": { # The sum of up and down votes given to the Answer. Can be negative.
"feature_type": "Value",
"dtype": "int32"
},
"Body": { # The body content of the Answer.
"feature_type": "Value",
"dtype": "string"
}
}
```
For Questions:
```
{
"Id": { # Unique ID given to the Question post
"feature_type": "Value",
"dtype": "int32"
},
"OwnerUserId": { # The UserID of the person who generated the Question on StackOverflow. -1 means NA.
"feature_type": "Value",
"dtype": "int32"
},
"CreationDate": { # The date the Question was generated. Follows standard datetime format.
"feature_type": "Value",
"dtype": "string"
},
"ClosedDate": { # The date the Question was generated. Follows standard datetime format. Can be NA.
"feature_type": "Value",
"dtype": "string"
},
"Score": { # The sum of up and down votes given to the Question. Can be negative.
"feature_type": "Value",
"dtype": "int32"
},
"Title": { # The title of the Question.
"feature_type": "Value",
"dtype": "string"
},
"Body": { # The body content of the Question.
"feature_type": "Value",
"dtype": "string"
}
}
```
For Tags:
```
{
"Id": { # ID of the Question the tag belongs to
"feature_type": "Value",
"dtype": "int32"
},
"Tag": { # The tag name
"feature_type": "Value",
"dtype": "string"
}
}
```
`
### Data Fields
For Answers:
-`Id`: Unique ID given to the Answer post
`OwnerUserId`: The UserID of the person who generated the Answer on StackOverflow. -1 means NA
"`CreationDate`": The date the Answer was generated. Follows standard datetime format.
"`ParentId`": Refers to the `Id` of the Question the Answer belong to.
"`Score`": The sum of up and down votes given to the Answer. Can be negative.
"`Body`": The body content of the Answer.
For Questions:
- `Id`: Unique ID given to the Question post.
- `OwnerUserId`: The UserID of the person who generated the Question on StackOverflow. -1 means NA.
- `CreationDate`: The date the Question was generated. Follows standard datetime format.
- `ClosedDate`: The date the Question was generated. Follows standard datetime format. Can be NA.
- `Score`: The sum of up and down votes given to the Question. Can be negative.
- `Title`: {The title of the Question.
- `Body`: The body content of the Question.
For Tags:
- `Id`: ID of the Question the tag belongs to.
- `Tag`: The tag name.
### Data Splits
The dataset has 3 splits:
- `Answers`
- `Questions`
- `Tags`
## Dataset Creation
### Curation Rationale
Datasets of all R questions and all Python questions are also available on Kaggle, but this dataset is especially useful for analyses that span many languages.
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
StackOverflow Users.
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
This data contains information that can identify individual users of StackOverflow. The information is self-reported.
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
StackOverflow answers are not guaranteed to be safe, secure, or correct. Some answers may purposefully be insecure as is done in this https://stackoverflow.com/a/35571883/5768407 answer from user [`zys`](https://stackoverflow.com/users/5259310/zys), where they show a solution to purposefully bypass Google Play store security checks. Such answers can lead to biased models that use this data and can further propogate unsafe and insecure programming practices.
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
All Stack Overflow user contributions are licensed under CC-BY-SA 3.0 with attribution required.
### Citation Information
The content is from Stack Overflow.
### Contributions
Thanks to [@ncoop57](https://github.com/ncoop57) for adding this dataset. |
false |
# Dataset Card for spc
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** http://opus.nlpl.eu/SPC.php
- **Repository:** None
- **Paper:** http://www.lrec-conf.org/proceedings/lrec2012/pdf/463_Paper.pdf
- **Leaderboard:** [More Information Needed]
- **Point of Contact:** [More Information Needed]
### Dataset Summary
[More Information Needed]
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
[More Information Needed]
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
[More Information Needed]
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@abhishekkrthakur](https://github.com/abhishekkrthakur) for adding this dataset. |
true |
# Dataset Card for [cedr]
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [GitHub](https://github.com/sag111/CEDR)
- **Repository:** [GitHub](https://github.com/sag111/CEDR)
- **Paper:** [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S1877050921013247)
- **Leaderboard:**
- **Point of Contact:** [@sag111](mailto:sag111@mail.ru)
### Dataset Summary
The Corpus for Emotions Detecting in Russian-language text sentences of different social sources (CEDR) contains 9410 comments labeled for 5 emotion categories (joy, sadness, surprise, fear, and anger).
Here are 2 dataset configurations:
- "main" - contains "text", "labels", and "source" features;
- "enriched" - includes all "main" features and "sentences".
Dataset with predefined train/test splits.
### Supported Tasks and Leaderboards
This dataset is intended for multi-label emotion classification.
### Languages
The data is in Russian.
## Dataset Structure
### Data Instances
Each instance is a text sentence in Russian from several sources with one or more emotion annotations (or no emotion at all).
An example for an instance from the dataset is shown below:
```
{
'text': 'Забавно как люди в возрасте удивляются входящим звонкам на мобильник)',
'labels': [0],
'source': 'twitter',
'sentences': [
[
{'forma': 'Забавно', 'lemma': 'Забавно'},
{'forma': 'как', 'lemma': 'как'},
{'forma': 'люди', 'lemma': 'человек'},
{'forma': 'в', 'lemma': 'в'},
{'forma': 'возрасте', 'lemma': 'возраст'},
{'forma': 'удивляются', 'lemma': 'удивляться'},
{'forma': 'входящим', 'lemma': 'входить'},
{'forma': 'звонкам', 'lemma': 'звонок'},
{'forma': 'на', 'lemma': 'на'},
{'forma': 'мобильник', 'lemma': 'мобильник'},
{'forma': ')', 'lemma': ')'}
]
]
}
```
Emotion label codes: {0: "joy", 1: "sadness", 2: "surprise", 3: "fear", 4: "anger"}
### Data Fields
The main configuration includes:
- text: the text of the sentence;
- labels: the emotion annotations;
- source: the tag name of the corresponding source
In addition to the above, the raw data includes:
- sentences: text tokenized and lemmatized with [udpipe](https://ufal.mff.cuni.cz/udpipe)
- 'forma': the original word form;
- 'lemma': the lemma of this word
### Data Splits
The dataset includes a set of train/test splits.
with 7528, and 1882 examples respectively.
## Dataset Creation
### Curation Rationale
The formed dataset of examples consists of sentences in Russian from several sources (blogs, microblogs, news), which allows creating methods to analyse various types of texts. The created methodology for building the dataset based on applying a crowdsourcing service can be used to expand the number of examples to improve the accuracy of supervised classifiers.
### Source Data
#### Initial Data Collection and Normalization
Data was collected from several sources: posts of the Live Journal social network, texts of the online news agency Lenta.ru, and Twitter microblog posts.
Only those sentences were selected that contained marker words from the dictionary of [the emotive vocabulary of the Russian language](http://lexrus.ru/default.aspx?p=2876). The authors manually formed a list of marker words for each emotion by choosing words from different categories of the dictionary.
In total, 3069 sentences were selected from LiveJournal posts, 2851 sentences from Lenta.Ru, and 3490 sentencesfrom Twitter. After selection, sentences were offered to annotators for labeling.
#### Who are the source language producers?
Russian-speaking LiveJournal and Tweeter users, and authors of news articles on the site lenta.ru.
### Annotations
#### Annotation process
Annotating sentences with labels of their emotions was performed with the help of [a crowdsourcing platform](https://yandex.ru/support/toloka/index.html?lang=en).
The annotators’ task was: “What emotions did the author express in the sentence?”. The annotators were allowed to put an arbitrary number of the following emotion labels: "joy", "sadness", "anger", "fear", and "surprise".
If the accuracy of an annotator on the control sentences (including the trial run) became less than 70%, or if the accuracy was less than 66% over the last six control samples, the annotator was dismissed.
Sentences were split into tasks and assigned to annotators so that each sentence was annotated at least three times. A label of a specific emotion was assigned to a sentence if put by more than half of the annotators.
#### Who are the annotators?
Only those of the 30% of the best-performing active users (by the platform’s internal rating) who spoke Russian and were over 18 years old were allowed into the annotation process. Moreover, before a platform user could be employed as an annotator, they underwent a training task, after which they were to mark 25 trial samples with more than 80% agreement compared to the annotation that the authors had performed themselves.
### Personal and Sensitive Information
The text of the sentences may contain profanity.
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Researchers at AI technology lab at NRC "Kurchatov Institute". See the author [list](https://www.sciencedirect.com/science/article/pii/S1877050921013247).
### Licensing Information
The GitHub repository which houses this dataset has an Apache License 2.0.
### Citation Information
If you have found our results helpful in your work, feel free to cite our publication. This is an updated version of the dataset, the collection and preparation of which is described here:
```
@article{sboev2021data,
title={Data-Driven Model for Emotion Detection in Russian Texts},
author={Sboev, Alexander and Naumov, Aleksandr and Rybka, Roman},
journal={Procedia Computer Science},
volume={190},
pages={637--642},
year={2021},
publisher={Elsevier}
}
```
### Contributions
Thanks to [@naumov-al](https://github.com/naumov-al) for adding this dataset. |
false |
The SETimes\_sr training corpus contains 86,726 Serbian tokens manually annotated on the levels of tokenisation, sentence segmentation, morphosyntactic tagging, lemmatisation, named entities and dependency syntax.
The dataset contains 3177 training samples, 395 validation samples and 319 test samples
across the respective data splits. Each sample represents a sentence and includes the following features:
sentence ID ('sent\_id'), sentence text ('text'), list of tokens ('tokens'), list of lemmas ('lemmas'),
list of MULTEXT-East tags ('xpos\_tags), list of UPOS tags ('upos\_tags'),
list of morphological features ('feats'), list of IOB tags ('iob\_tags') and list of universal dependencies ('uds').
Three dataset configurations are available, namely 'ner', 'upos', and 'ud', with the corresponding features
encoded as class labels. If the configuration is not specified, it defaults to 'ner'.
If you use this dataset in your research, please cite the following paper:
```
@inproceedings{samardzic-etal-2017-universal,
title = "{U}niversal {D}ependencies for {S}erbian in Comparison with {C}roatian and Other {S}lavic Languages",
author = "Samard{\v{z}}i{\'c}, Tanja and
Starovi{\'c}, Mirjana and
Agi{\'c}, {\v{Z}}eljko and
Ljube{\v{s}}i{\'c}, Nikola",
booktitle = "Proceedings of the 6th Workshop on {B}alto-{S}lavic Natural Language Processing",
month = apr,
year = "2017",
address = "Valencia, Spain",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/W17-1407",
doi = "10.18653/v1/W17-1407",
pages = "39--44",
}
``` |
true |
# Dataset Summary
This dataset contains natural language descriptions of symptoms labeled with 22 corresponding diagnoses. `Gretel/symptom_to_diagnosis` provides 1065 symptom descriptions in the English language labeled with 22 diagnoses, focusing on fine-grained single-domain diagnosis.
## Data Fields
Each row contains the following fields:
* `input_text` : A string field containing symptoms
* `output_text` : A string field containing a diagnosis
Example:
```
{
"output_text": "drug reaction",
"input_text": "I've been having headaches and migraines, and I can't sleep. My whole body shakes and twitches. Sometimes I feel lightheaded."
}
```
## Diagnoses
This table contains the count of each diagnosis in the train and test splits.
| | Diagnosis | train.jsonl | test.jsonl |
|---:|:--------------------------------|--------------:|-------------:|
| 0 | drug reaction | 40 | 8 |
| 1 | allergy | 40 | 10 |
| 2 | chicken pox | 40 | 10 |
| 3 | diabetes | 40 | 10 |
| 4 | psoriasis | 40 | 10 |
| 5 | hypertension | 40 | 10 |
| 6 | cervical spondylosis | 40 | 10 |
| 7 | bronchial asthma | 40 | 10 |
| 8 | varicose veins | 40 | 10 |
| 9 | malaria | 40 | 10 |
| 10 | dengue | 40 | 10 |
| 11 | arthritis | 40 | 10 |
| 12 | impetigo | 40 | 10 |
| 13 | fungal infection | 39 | 9 |
| 14 | common cold | 39 | 10 |
| 15 | gastroesophageal reflux disease | 39 | 10 |
| 16 | urinary tract infection | 39 | 9 |
| 17 | typhoid | 38 | 9 |
| 18 | pneumonia | 37 | 10 |
| 19 | peptic ulcer disease | 37 | 10 |
| 20 | jaundice | 33 | 7 |
| 21 | migraine | 32 | 10 |
## Data Splits
The data is split to 80% train (853 examples, 167kb) and 20% test (212 examples, 42kb).
## Dataset Creation
Data was filtered to remove unwanted categories and updated using an LLM to create language more consistent with how a patient would describe symptoms in natural language to a doctor.
## Source Data
This dataset was adapted based on the [Symptom2Disease](https://www.kaggle.com/datasets/niyarrbarman/symptom2disease) dataset from Kaggle.
## Personal and Sensitive Information
The symptoms in this dataset were modified from their original format using an LLM and do not contain personal data.
## Limitations
This dataset is licensed Apache 2.0 and free for use. |
false | # Dataset Card for the args.me corpus
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Dataset Usage](#dataset-usage)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://zenodo.org/record/4139439
- **Repository:** https://git.webis.de/code-research/arguana/args/args-framework
- **Paper:** [Building an Argument Search Engine for the Web](https://webis.de/downloads/publications/papers/wachsmuth_2017f.pdf)
- **Leaderboard:** https://touche.webis.de/
- **Point of Contact:** [Webis Group](https://webis.de/people.html)
### Dataset Summary
The args.me corpus (version 1.0, cleaned) comprises 382 545 arguments crawled from four debate portals in the middle of 2019. The debate portals are Debatewise, IDebate.org, Debatepedia, and Debate.org. The arguments are extracted using heuristics that are designed for each debate portal.
### Dataset Usage
```python
import datasets
args = datasets.load_dataset('cakiki/args_me', 'corpus', streaming=True)
args_iterator = iter(args)
for arg in args_iterator:
print(args['conclusion'])
print(args['id'])
print(args['argument'])
print(args['stance'])
break
```
### Supported Tasks and Leaderboards
Document Retrieval, Argument Retrieval for Controversial Questions
### Languages
The args.me corpus is monolingual; it only includes English (mostly en-US) documents.
## Dataset Structure
### Data Instances
#### Corpus
```
{'conclusion': 'Science is the best!',
'id': 'd6517702-2019-04-18T12:36:24Z-00000-000',
'argument': 'Science is aright I guess, but Physical Education (P.E) is better. Think about it, you could sit in a classroom for and hour learning about molecular reconfiguration, or you could play football with your mates. Why would you want to learn about molecular reconfiguration anyway? I think the argument here would be based on, healthy mind or healthy body. With science being the healthy mind and P.E being the healthy body. To work this one out all you got to do is ask Steven Hawkins. Only 500 words',
'stance': 'CON'}
```
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[Creative Commons Attribution 4.0 International (CC BY 4.0)](https://creativecommons.org/licenses/by/4.0/)
### Citation Information
```
@dataset{yamen_ajjour_2020_4139439,
author = {Yamen Ajjour and
Henning Wachsmuth and
Johannes Kiesel and
Martin Potthast and
Matthias Hagen and
Benno Stein},
title = {args.me corpus},
month = oct,
year = 2020,
publisher = {Zenodo},
version = {1.0-cleaned},
doi = {10.5281/zenodo.4139439},
url = {https://doi.org/10.5281/zenodo.4139439}
}
```
|
false |
The hr500k training corpus contains 506,457 Croatian tokens manually annotated on the levels of tokenisation, sentence segmentation, morphosyntactic tagging, lemmatisation, named entities and dependency syntax.
On the sentence level, the dataset contains 20159 training samples, 1963 validation samples and 2672 test samples
across the respective data splits. Each sample represents a sentence and includes the following features:
sentence ID ('sent\_id'), sentence text ('text'), list of tokens ('tokens'), list of lemmas ('lemmas'),
list of MULTEXT-East tags ('xpos\_tags), list of UPOS tags ('upos\_tags'), list of morphological features ('feats'),
and list of IOB tags ('iob\_tags'). A subset of the data also contains universal dependencies ('ud') and consists of
7498 training samples, 649 validation samples, and 742 test samples.
Three dataset configurations are available, namely 'ner', 'upos', and 'ud', with the corresponding features
encoded as class labels. If the configuration is not specified, it defaults to 'ner'.
If you use this dataset in your research, please cite the following paper:
```
Bibtex @InProceedings{LJUBEI16.340,
author = {Nikola Ljubešić and Filip Klubička and Željko Agić and Ivo-Pavao Jazbec},
title = {New Inflectional Lexicons and Training Corpora for Improved Morphosyntactic Annotation of Croatian and Serbian},
booktitle = {Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016)},
year = {2016},
month = {may},
date = {23-28},
location = {Portorož, Slovenia},
editor = {Nicoletta Calzolari (Conference Chair) and Khalid Choukri and Thierry Declerck and Sara Goggi and Marko Grobelnik and Bente Maegaard and Joseph Mariani and Helene Mazo and Asuncion Moreno and Jan Odijk and Stelios Piperidis},
publisher = {European Language Resources Association (ELRA)},
address = {Paris, France},
isbn = {978-2-9517408-9-1},
language = {english}
}
``` |
false |
# Dataset Card for swedish_medical_ner
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** https://github.com/olofmogren/biomedical-ner-data-swedish
- **Paper:** [Named Entity Recognition in Swedish Health Records with Character-Based Deep Bidirectional LSTMs](https://aclanthology.org/W16-5104.pdf)
- **Point of Contact:** [Olof Mogren](olof@mogren.one)
### Dataset Summary
SwedMedNER is Named Entity Recognition dataset on medical text in Swedish. It consists three subsets which are in turn derived from three different sources respectively: the Swedish Wikipedia (a.k.a. wiki), Läkartidningen (a.k.a. lt), and 1177 Vårdguiden (a.k.a. 1177). While the Swedish Wikipedia and Läkartidningen subsets in total contains over 790000 sequences with 60 characters each, the 1177 Vårdguiden subset is manually annotated and contains 927 sentences, 2740 annotations, out of which 1574 are _disorder and findings_, 546 are _pharmaceutical drug_, and 620 are _body structure_.
Texts from both Swedish Wikipedia and Läkartidningen were automatically annotated using a list of medical seed terms. Sentences from 1177 Vårdguiden were manuually annotated.
### Supported Tasks and Leaderboards
Medical NER.
### Languages
Swedish (SV).
## Dataset Structure
### Data Instances
Annotated example sentences are shown below:
```
( Förstoppning ) är ett vanligt problem hos äldre.
[ Cox-hämmare ] finns även som gel och sprej.
[ Medicinen ] kan också göra att man blöder lättare eftersom den påverkar { blodets } förmåga att levra sig.
```
Tags are as follows:
- Prenthesis, (): Disorder and Finding
- Brackets, []: Pharmaceutical Drug
- Curly brackets, {}: Body Structure
Data example:
```
In: data = load_dataset('./datasets/swedish_medical_ner', "wiki")
In: data['train']:
Out:
Dataset({
features: ['sid', 'sentence', 'entities'],
num_rows: 48720
})
In: data['train'][0]['sentence']
Out: '{kropp} beskrivs i till exempel människokroppen, anatomi och f'
In: data['train'][0]['entities']
Out: {'start': [0], 'end': [7], 'text': ['kropp'], 'type': [2]}
```
### Data Fields
- `sentence`
- `entities`
- `start`: the start index
- `end`: the end index
- `text`: the text of the entity
- `type`: entity type: Disorder and Finding (0), Pharmaceutical Drug (1) or Body Structure (2)
### Data Splits
In the original paper, its authors used the text from Läkartidningen for model training, Swedish Wikipedia for validation, and 1177.se for the final model evaluation.
## Dataset Creation
### Curation Rationale
### Source Data
- Swedish Wikipedia;
- Läkartidningen - contains articles from the Swedish journal for medical professionals;
- 1177.se - a web site provided by the Swedish public health care authorities, containing information, counselling, and other health-care services.
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
- A list of seed terms was extracted using SweMeSH and SNOMED CT;
- The following predefined categories was used for the extraction: disorder & finding (sjukdom & symtom), pharmaceutical drug (läkemedel) and body structure (kroppsdel)
- For _Swedish Wikipedia_, an initial list of medical domain articles were selected manually. These source articles as well as their linked articles were downloaded and automatically annotated by finding the aforementioned seed terms with a context window of 60 characters;
- Articles from the _Läkartidningen_ corpus were downloaded and automatically annotated by finding the aforementioned seed terms with a context window of 60 characters;
- 15 documents from _1177.se_ were downloaded in May 2016 and then manually annotated with the seed terms as support, resulting 2740 annotations.
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
- Simon Almgren, simonwalmgren@gmail.com
- Sean Pavlov, sean.pavlov@gmail.com
- Olof Mogren, olof@mogren.one
Chalmers University of Technology
### Licensing Information
This dataset is released under the [Creative Commons Attribution-ShareAlike 4.0 International Public License (CC BY-SA 4.0)](http://creativecommons.org/licenses/by-sa/4.0/).
### Citation Information
```bibtex
@inproceedings{almgrenpavlovmogren2016bioner,
title={Named Entity Recognition in Swedish Medical Journals with Deep Bidirectional Character-Based LSTMs},
author={Simon Almgren, Sean Pavlov, Olof Mogren},
booktitle={Proceedings of the Fifth Workshop on Building and Evaluating Resources for Biomedical Text Mining (BioTxtM 2016)},
pages={1},
year={2016}
}
```
### Contributions
Thanks to [@bwang482](https://github.com/bwang482) for adding this dataset. |
false | The dataset contains 7432 training samples, 1164 validation samples and 893 test samples.
Each sample represents a sentence and includes the following features: sentence ID ('sent\_id'),
list of tokens ('tokens'), list of lemmas ('lemmas'),
list of Multext-East tags ('xpos\_tags), list of UPOS tags ('upos\_tags'), list of morphological features ('feats'),
list of IOB tags ('iob\_tags'), and list of universal dependency tags ('uds'). Three dataset configurations are
available, where the corresponding features are encoded as class labels: 'ner', 'upos', and 'ud'. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.