Abstract
The use of Massive Online Open Courses (MOOCs) has been noticeably increased in recent times, especially after the COVID-19 pandemic. In the absence of one-to-one interaction with the students, the instructors are no longer able to understand the demands of their students in an intrinsic way. To overcome this problem, the MOOC platforms provide a discussion forum in which students can share their thoughts and problems about the course. The instructors must closely monitor the performance of their students so that they can improve their teaching methodology to enhance the students' understanding. The instructors must go through the long chats in the discussion forums to identify specific problem areas faced by students. In this study, we propose a method that first categorizes discussion threads into topics and subtopics with the help of topic modeling and then performs sentiment analysis on comments to identify the sentiment of the posts. The primary objective of the study is to facilitate the instructors so that they can improve their teaching methodology, thus enhancing the understanding level of the students.