InsightQA: Descriptive Question and Answer Generation

Present research largely concentrates on the generation and evaluation of objective-type questions, with significantly less focus on automating the creation of subjective questions and answers. As the demand for contextually relevant question answer generation grows, the manual task of setting of subjective question papers and assessment of their descriptive responses becomes increasingly burdensome and time-consuming. As a consequence of this, the educators are unable to dedicate time to more crucial tasks of strategic and engaging educational activities. Also, the existing data sets for descriptive question answer pairs contain very short length answers such as answer in two to three words or very short single sentence answers. To tackle these challenges, we have created our own data set labelled DescQA which includes question answer pairs from diverse educational domains and an Advanced Language Model (ALM) that automates the generation of contextually appropriate descriptive questions and answers. Utilizing cutting-edge natural language processing (NLP) techniques and a diverse dataset specifically designed for this purpose, our model accurately interprets educational materials to produce contextually relevant descriptive questions and answers. This initiative aims to reduce the workload on paper setters, enabling them to prioritize personalized instruction and student engagement. By generating more contextually relevant descriptive questions and answers, the ALM aspires to improve the learning experience for both educators and learners.