Recommending Question-Answers for Enriching Textbooks

摘要

The students often use the community question-answering (cQA) systems along with textbooks to gain more knowledge. The millions of question-answers (QA) already accessible in these cQA forums. The huge amount of QA makes it hard for the students to go through all possible QA for a better understanding of the concepts. To address this issue, this paper provides a technological solution "Topic-based text enrichment process" for a textbook with relevant QA pairs from cQA. We used techniques of natural language processing, topic modeling, and data mining to extract the most relevant QA sets and corresponding links of cQA to enrich the textbooks. This work provides all the relevant Q As for the important topics of the textbook to the students, therefore it helps them to learn the concept more quickly. Experiments were carried out on a variety of textbooks such as Pattern Recognition and Machine Learning by Christopher M Bishop, Data Mining Concepts & techniques by Jiawei Han, Information Theory, Inference, & Learning Algorithms by David J.C. MacKay, and National Council of Educational Research and Training (NCERT). The results prove that we are effective in learning enhancement by enhancing the textbooks on various subjects and across different grades with the relevant QA pairs using automated techniques. We also present the results of quiz sessions which were conducted to evaluate the effectiveness of the proposed model in establishing relevance to learning among students.