Update README.md
Browse files
README.md
CHANGED
|
@@ -8,52 +8,76 @@ tags:
|
|
| 8 |
task_categories:
|
| 9 |
- text-classification
|
| 10 |
- text-generation
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
# Dataset Card for Dataset Name
|
| 14 |
|
| 15 |
## Dataset Description
|
| 16 |
|
| 17 |
-
- **Homepage: (TODO) https://docs.ryokoai.com/docs/training/dataset#Fandom22K
|
| 18 |
-
- **Repository: (TODO) https://github.com/RyokoAI/BigKnow2022
|
| 19 |
- **Paper: N/A**
|
| 20 |
- **Leaderboard: N/A**
|
| 21 |
- **Point of Contact: Ronsor Labs <ronsor@ronsor.com>**
|
| 22 |
|
| 23 |
### Dataset Summary
|
| 24 |
|
| 25 |
-
Fandom23K is a dataset composed of articles scraped from approximately 23,665 Fandom.com wikis between March 14 and March 18, 2023.
|
| 26 |
It is a subset of the upcoming BigKnow2022 dataset.
|
| 27 |
|
| 28 |
### Supported Tasks and Leaderboards
|
| 29 |
|
| 30 |
-
|
|
|
|
|
|
|
| 31 |
|
| 32 |
### Languages
|
| 33 |
|
| 34 |
* English
|
|
|
|
| 35 |
|
| 36 |
## Dataset Structure
|
| 37 |
|
| 38 |
### Data Instances
|
| 39 |
|
| 40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
|
| 42 |
### Data Fields
|
| 43 |
|
| 44 |
* **text**: the actual article text
|
| 45 |
* **title**: the article title
|
| 46 |
-
* **tag**: source tag, in the format fandom.*<wiki name>*
|
| 47 |
|
| 48 |
### Data Splits
|
| 49 |
|
| 50 |
-
|
| 51 |
|
| 52 |
## Dataset Creation
|
| 53 |
|
| 54 |
### Curation Rationale
|
| 55 |
|
| 56 |
-
|
|
|
|
|
|
|
| 57 |
|
| 58 |
### Source Data
|
| 59 |
|
|
@@ -64,28 +88,29 @@ This list was gathered in early January 2023, despite the actual wiki content be
|
|
| 64 |
|
| 65 |
#### Who are the source language producers?
|
| 66 |
|
| 67 |
-
The
|
| 68 |
|
| 69 |
### Annotations
|
| 70 |
|
| 71 |
#### Annotation process
|
| 72 |
|
| 73 |
-
Wiki names and article titles were collected alongside the article text. Other than that, no annotation was performed.
|
| 74 |
|
| 75 |
#### Who are the annotators?
|
| 76 |
|
| 77 |
-
|
| 78 |
|
| 79 |
### Personal and Sensitive Information
|
| 80 |
|
| 81 |
The dataset was collected from public wiki data. As a result, we do not believe
|
| 82 |
-
it
|
| 83 |
|
| 84 |
## Considerations for Using the Data
|
| 85 |
|
| 86 |
### Social Impact of Dataset
|
| 87 |
|
| 88 |
-
|
|
|
|
| 89 |
|
| 90 |
### Discussion of Biases
|
| 91 |
|
|
@@ -94,7 +119,8 @@ Additionally, this dataset was not filtered at all. We recommmend its usage for
|
|
| 94 |
|
| 95 |
### Other Known Limitations
|
| 96 |
|
| 97 |
-
|
|
|
|
| 98 |
|
| 99 |
## Additional Information
|
| 100 |
|
|
|
|
| 8 |
task_categories:
|
| 9 |
- text-classification
|
| 10 |
- text-generation
|
| 11 |
+
pretty_name: Fandom23K Wikis
|
| 12 |
+
size_categories:
|
| 13 |
+
- 10M<n<100M
|
| 14 |
---
|
| 15 |
|
| 16 |
# Dataset Card for Dataset Name
|
| 17 |
|
| 18 |
## Dataset Description
|
| 19 |
|
| 20 |
+
- **Homepage: (TODO) https://docs.ryokoai.com/docs/training/dataset#Fandom22K**
|
| 21 |
+
- **Repository: (TODO) https://github.com/RyokoAI/BigKnow2022**
|
| 22 |
- **Paper: N/A**
|
| 23 |
- **Leaderboard: N/A**
|
| 24 |
- **Point of Contact: Ronsor Labs <ronsor@ronsor.com>**
|
| 25 |
|
| 26 |
### Dataset Summary
|
| 27 |
|
| 28 |
+
Fandom23K is a dataset composed of 15,616,749 articles scraped from approximately 23,665 Fandom.com wikis between March 14 and March 18, 2023.
|
| 29 |
It is a subset of the upcoming BigKnow2022 dataset.
|
| 30 |
|
| 31 |
### Supported Tasks and Leaderboards
|
| 32 |
|
| 33 |
+
This dataset is primarily intended for unsupervised training of text generation models; however, it may be useful for other purposes.
|
| 34 |
+
|
| 35 |
+
* text-classification
|
| 36 |
|
| 37 |
### Languages
|
| 38 |
|
| 39 |
* English
|
| 40 |
+
* Potentially other languages in much smaller quantities.
|
| 41 |
|
| 42 |
## Dataset Structure
|
| 43 |
|
| 44 |
### Data Instances
|
| 45 |
|
| 46 |
+
```json
|
| 47 |
+
{
|
| 48 |
+
"tag": "fandom.wikia2011",
|
| 49 |
+
"text": "# Add Your Wiki's Highlights\n\nWrite the text of your article here!-_-\n\n",
|
| 50 |
+
"title": "Add Your Wiki's Highlights"
|
| 51 |
+
}
|
| 52 |
+
{
|
| 53 |
+
"tag": "fandom.wikia2011",
|
| 54 |
+
"text": "# Add Your Wiki's Highlights!\n\nWikia wants to hear from you! What significant milestones did your wiki experience in 2011? What cool things did the community try out?\nCreate a page for the wiki you're most active on! Be sure to add it to the Entertainment, Gaming, or Lifestyle categories so it shows up in the right place!\n\n",
|
| 55 |
+
"title": "Add Your Wiki's Highlights!"
|
| 56 |
+
}
|
| 57 |
+
{
|
| 58 |
+
"tag": "fandom.wikia2011",
|
| 59 |
+
"text": "# Assassins Creed Wiki 2011\n\nIn 2011, Assassin's Creed Wiki tested new Wikia features such as Message Wall, Chat, and New Layouts.\n\n",
|
| 60 |
+
"title": "Assassins Creed Wiki 2011"
|
| 61 |
+
}
|
| 62 |
+
```
|
| 63 |
|
| 64 |
### Data Fields
|
| 65 |
|
| 66 |
* **text**: the actual article text
|
| 67 |
* **title**: the article title
|
| 68 |
+
* **tag**: text source tag, in the following format: fandom.*<wiki name>*
|
| 69 |
|
| 70 |
### Data Splits
|
| 71 |
|
| 72 |
+
No splitting of the data was performed. This dataset is intended for unsupervised training.
|
| 73 |
|
| 74 |
## Dataset Creation
|
| 75 |
|
| 76 |
### Curation Rationale
|
| 77 |
|
| 78 |
+
Fandom23K provides an up-to-date corpus containing pop culture and media information spanning a variety of interests and
|
| 79 |
+
hobbies. Previous datasets containing such information are either part of a large and harder-to-handle whole, such as
|
| 80 |
+
Common Crawl, do not provide enough variety, or are simply outdated.
|
| 81 |
|
| 82 |
### Source Data
|
| 83 |
|
|
|
|
| 88 |
|
| 89 |
#### Who are the source language producers?
|
| 90 |
|
| 91 |
+
The contributors of each wiki.
|
| 92 |
|
| 93 |
### Annotations
|
| 94 |
|
| 95 |
#### Annotation process
|
| 96 |
|
| 97 |
+
Wiki names and article titles were collected alongside the article text. Other than that automated process, no annotation was performed.
|
| 98 |
|
| 99 |
#### Who are the annotators?
|
| 100 |
|
| 101 |
+
There were no human annotators.
|
| 102 |
|
| 103 |
### Personal and Sensitive Information
|
| 104 |
|
| 105 |
The dataset was collected from public wiki data. As a result, we do not believe
|
| 106 |
+
it should contain any PII and did not inspect it further.
|
| 107 |
|
| 108 |
## Considerations for Using the Data
|
| 109 |
|
| 110 |
### Social Impact of Dataset
|
| 111 |
|
| 112 |
+
This dataset is intended to be useful for anyone who wishes to train a model to generate "more entertaining" content requiring
|
| 113 |
+
knowledge of popular culture or a particular niche.
|
| 114 |
|
| 115 |
### Discussion of Biases
|
| 116 |
|
|
|
|
| 119 |
|
| 120 |
### Other Known Limitations
|
| 121 |
|
| 122 |
+
This dataset is based on a list of active wikis from January 2023, even though the actual wiki content may be more recent. Additionally,
|
| 123 |
+
smaller yet still active wikis may have been excluded.
|
| 124 |
|
| 125 |
## Additional Information
|
| 126 |
|