metadata

configs:
  - config_name: balanced
    data_files:
      - split: train
        path: balanced/train-*
  - config_name: unbalanced
    data_files:
      - split: train
        path: unbalanced/train-*
dataset_info:
  - config_name: balanced
    features:
      - name: rating
        dtype: int64
      - name: review id
        dtype: int64
      - name: user id
        dtype: int64
      - name: book id
        dtype: int64
      - name: review
        dtype: string
    splits:
      - name: train
        num_bytes: 117555389
        num_examples: 156506
    download_size: 62785913
    dataset_size: 117555389
  - config_name: unbalanced
    features:
      - name: rating
        dtype: int64
      - name: review id
        dtype: int64
      - name: user id
        dtype: int64
      - name: book id
        dtype: int64
      - name: review
        dtype: string
    splits:
      - name: train
        num_bytes: 401066853
        num_examples: 510599
    download_size: 212079509
    dataset_size: 401066853
license: unknown
task_categories:
  - text-classification
language:
  - ar
tags:
  - Sentiment Analysis
pretty_name: BRAD
size_categories:
  - 100K<n<1M

Dataset Card for "BRAD: Books Reviews in Arabic Dataset"

Paper:

Elnagar A. and Einea O. 'BRAD 1.0: Book reviews in Arabic dataset'. 2016 IEEE/ACS 13th International Conference of Computer Systems and Applications (AICCSA), pp. 1-8, Nov 2016. DOI: 10.1109/AICCSA.2016.7945800.