Relevancy prediction of micro-blog questions in an educational setting

Micro-blogging has become increasingly popular in recent years. Using micro-blogging in a large classroom could be beneficial for learning. However, sometimes addressing the large number of posts could be cumbersome to a reader who has only limited time in a classroom. We propose a novel solution for predicting the relevancy of a question asked in a class by looking at the questions asked in previous semesters, the similarity of the question to the lecture material, as well as a set of question features such as the number of students’ votes, number of replies, the length of the question, and whether it was asked anonymously. To identify similar questions asked previously, topic modeling and feature selection are used. Empirical results show that topic modeling leads to better prediction performance score as compared to feature selection. The similarity of the question and its corresponding lecture material further improves the relevancy prediction of the questions.