metadata

configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: word
      dtype: string
    - name: tag
      dtype: string
  splits:
    - name: train
      num_bytes: 3474380
      num_examples: 168080
  download_size: 1047962
  dataset_size: 3474380
license: cc-by-sa-3.0
language:
  - ar
tags:
  - Named Entity Recognition
task_categories:
  - token-classification
size_categories:
  - 100K<n<1M

Dataset Card for "NewsFANE: Newswire-based Fine-grained Arabic Named Entity Corpus"

Paper:

Fahd Alotaibi and Mark Lee. 2014. A Hybrid Approach to Features Representation for Fine-grained Arabic Named Entity Recognition. In Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers, pages 984–995, Dublin, Ireland. Dublin City University and Association for Computational Linguistics.