UDDIPOK: Reading Comprehension Based Question Answering Dataset in Bangla Language

Published: 2 December 2022| Version 3 | DOI: 10.17632/s9pb3h2cjy.3
Tanjim Taharat Aurpa,


The urgency of the online education system can be understood during the COVID-19 pandemic. The entire world, including Bangladesh, moved to online education and focused on developing automated systems. This dataset is created for developing automatic reading comprehension systems in the Bangla language. We have also provided the English Translation of this dataset. The dataset contains the following parameters: 1. A real-time and long passage (context) in the Bengali language is collected from different Bangla articles, novels, biography, etc. 2. Questions are generated based on the given passages. 3. Answers' text of the given passage and question. 4. The start index of the answer in the passage.



Jahangirnagar University


Natural Language Processing, Bangladesh, Deep Learning