---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
dataset_info:
  features:
  - name: Id
    dtype: int64
  - name: Platform
    dtype: string
  - name: Comment
    dtype: string
  - name: Majority_Label
    dtype: string
  - name: Agreement
    dtype: float64
  - name: NumOfJudgementUsed
    dtype: int64
  - name: Total_Judgement
    dtype: int64
  - name: Vulgar:V/HateSpeech:HS/None:-
    dtype: string
  splits:
  - name: train
    num_bytes: 1184762
    num_examples: 4000
  download_size: 561173
  dataset_size: 1184762
license: apache-2.0
task_categories:
- text-classification
language:
- ar
size_categories:
- 1K<n<10K
tags:
- Offensive Language Detection
---
# Dataset Card for "Arabic_Offensive_Comment_Detection"

## Paper:
Shammur Absar Chowdhury, Hamdy Mubarak, Ahmed Abdelali, Soon-gyo Jung, Bernard J. Jansen, and Joni Salminen. 2020. A Multi-Platform Arabic News Comment Dataset for Offensive Language Detection. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 6203–6212, Marseille, France. European Language Resources Association.