Building a benchmark dataset for the Kurdish news question answering

Ari M Saeed1

  • 1Computer Science Department, College of Science, University of Halabja, Kurdistan Region, Halabja, Iraq.

Data in Brief
|September 24, 2024
PubMed
Summary

This study introduces the Kurdish News Question Answering Dataset (KNQAD), a valuable resource for natural language processing research. The dataset features 15,002 Kurdish news paragraphs with manually created question-answer pairs, aiding in the development of Kurdish language AI models.