Providing accurate course/video recommendations in E-Learning environment using association rule mining and collaborative filtering

Abstract

Internet has huge number of learning resources, reason why students fail to take benefit of those is because they don't know where to look for resources, and more importantly which of these will be ideal for their respective academics. To provide related content to students most of the institutes uses different E-Learning solutions which acts as a repository of learning resources for students.
These E-learning solutions often don't provide personalized recommendations to users. We propose an E-Learning solution which provides users with recommendations based on his/her preferences and content consumed by similar students, further more system we propose provide all the facilities like course sharing between two universities, online tests, analytics etc. in one software. Collaborative filtering and its modifications is one of the most commonly used recommendation algorithm. Collaborative Filtering find people with similar interests, analyze their behavior derived from their ratings, and recommend target user the same items. As online social networks are growing, users can now make friends, share thoughts, images etc. on the Internet and express different level of trust on their web friends. Recommendations generated by the trusted friends are more relevant than other users. This paper proposes a video recommendation system that generates recommendations from the collaboration of trusted friends of the target user and uses association rule mining to capture current trends of users in the network.