Smart Approach for Multimedia Resources Recommendation Using Collabortive Filtering

A recommendation system predicts the information by filtering it on the basis of user rating and likes. Educational videos are common source of learning now a days. Education video recommendation systems are getting famous as they are dependent on multimedia resources as a tool to enhance E-learning process. Most of these Multimedia E-learning platforms lack real time accuracy in terms of relevance and ranking as per a certain learner's requirements. This study addresses the relevance Multimedia recommendation problem using Machine Learning algorithm and aims to improve the ranking through collaborative filtering approach which is applied on a web-based prototype of Multimedia based Learning Management System (LMS). By using DB Scan Algorithm, we are able to classify each Multimedia resource correctly. The present work improves the ranking results of recommended Multimedia resources in e-Learning Environment and proposed top N recommendations list to learner. The recommended list is calculated on user interest, likes and views. The proposed approach is evaluated and provided better accuracy of 96% and MAP of 88% when compared to existing video recommendation systems.