README.md

Arabic Reading Comprehension Dataset (ARCD)

The format of both ARCD and SQuAD follow (taken from https://github.com/facebookresearch/DrQA):

file.json
├── "data"
│   └── [i]
│       ├── "paragraphs"
│       │   └── [j]
│       │       ├── "context": "paragraph text"
│       │       └── "qas"
│       │           └── [k]
│       │               ├── "answers"
│       │               │   └── [l]
│       │               │       ├── "answer_start": N
│       │               │       └── "text": "answer"
│       │               ├── "id": "<uuid>"
│       │               └── "question": "paragraph question?"
│       └── "title": "document id"
└── "version": 1.1

Arabic Translated Stanford Question Answering Dataset (Arabic-SQuAD)