Abstract
Massive Open Online Courses (MOOCs) scale up their class size by eliminating synchronous teaching and the need for students and instructors to be co-located. Yet, the very characteristics that enable scalability of MOOCs also bring a significant challenge to its teaching. In particular, when students discuss problems in the forum, instructors may not be available to view and provide suggestion recently. Especially, when students ask for knowledge-based questions, which can be addressed in one of the videos in the course, it takes a long time for them to review all the course videos.
With an AI-based recommendation model, content of videos and post discussions can be extracted from the MOOC system. The model then is able to address the relations between the discussion and the video recourse, and recommend video clips to students post as a result. Such a recommendation system is able to significantly help students enhance learning and review in the MOOC study.
Resources
- https://www.aclweb.org/anthology/I17-1088.pdf
- http://www.liangmingpan.com/files/publications/IEEE_Access.pdf
- https://www.comp.nus.edu.sg/~kanmy/papers/muir-framework-cross.pdf
Members
- Hu Hengchang (Graduate Student)
- Zhang Tianyang (FYP Student)
- Min-Yen Kan (Advisor and Professor)