Datasets:

samanjoy2
/

multiconer_v1

pretty_name: MultiCoNER v1
license: cc-by-4.0
task_categories:
  - token-classification
language:
  - bn
  - de
  - en
  - es
  - fa
  - hi
  - ko
  - nl
  - ru
  - tr
  - zh
tags:
  - named-entity-recognition
  - ner
  - multilingual
  - multiconer
  - semeval-2022

MultiCoNER v1

MultiCoNER v1 is a multilingual dataset for complex named entity recognition.

This repository is an unofficial convenience mirror created to make the publicly available dataset easier to access through Hugging Face. I did not create or own this dataset. All credit belongs to the original MultiCoNER authors and the organizers of SemEval-2022 Task 11.

Entity types

The dataset uses BIO annotations for six entity types: Person (PER), Location (LOC), Group (GRP), Corporation (CORP), Product (PROD), and Creative Work (CW).

Data format

The files are provided in CoNLL format. Each line contains a token and its named entity label, while empty lines separate samples.

Download

from huggingface_hub import snapshot_download

dataset_path = snapshot_download(
    repo_id="samanjoy2/multiconer_v1",
    repo_type="dataset"
)

License

The original dataset is distributed under the Creative Commons Attribution 4.0 International license.

Citation

@inproceedings{malmasi-etal-2022-multiconer,
    title = "{M}ulti{C}o{NER}: A Large-scale Multilingual Dataset for Complex Named Entity Recognition",
    author = "Malmasi, Shervin and Fang, Anjie and Fetahu, Besnik and Kar, Sudipta and Rokhlenko, Oleg",
    booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
    year = "2022",
    pages = "3798--3809",
    url = "https://aclanthology.org/2022.coling-1.334/"
}

This repository is only a convenience mirror. Please cite the original dataset authors.