Development of a Domain-Specific Educational Corpus for Adaptive Intelligent Question-Answering Systems

Résumé fourni par la source

Developing a corpus for a specific domain is one of the major steps in enabling an intelligent system that can provide accurate and personalized responses, particularly in the field of academic advising. In adaptive learning, the specialized educational corpora allow AI models to generate responses that are not only correct but also personalized to individual students' learning preferences, questions, as well as academic goals. This chapter presents the development of question–answering corpus designed to support high school students in their academic and career decision-making using English language. The corpus consists of 10,318 manually collected and annotated question–answer pairs which collected from various educational sources, including students, parents, and educational websites. The dataset was analyzed to understand the distribution of question types and was processed using several preprocessing techniques, such as data cleaning, normalization, tokenization, and sequence padding.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.