# Question Answering Corpus

Extractive question answering dataset derived from Wikipedia articles.
Each record contains a context paragraph, a question, and the answer span.

Source: https://huggingface.co/datasets/rajpurkar/squad

## Dataset Structure
- Number of rows: 100,000+
- Columns: id, title, context, question, answers

## Usage
This dataset is intended for reading comprehension model training.