MOOC-FRS: A new fusion recommender system for MOOCs
Yunchou Li, Hongkun Li · 2017
The popularity of MOOCs has broken the limitation on time and space in traditional education. It has become an important source of accessing high quality resources and systematic education for the majority of learners. MOOCs has been greatly developed in recent years, followed by the surge in the number of users and courses, which making the learners exposed to a wild variety of courses and difficult to make the choice. Therefore, building a practical recommender system for MOOCs platforms has become increasing important. In this paper, we summarize the representative characteristics of MOOCs platform and proposed a new fusion recommender system called MOOC-FRS. We design a new metric to measure the relevance between user and courses based on the special user behavior in MOOCs platform. The recommender system is mainly built on course-based collaborative recommendation and we consider plenty of practical problems like the new user and new courses. Besides, we also propose a completely new adaptation to the recommender system called correlated pattern based recommendation which is able to combine the advantage of both user-based clustering and course-based clustering.