Interaction based collaborative filtering for personal elearning environment

Abstract

E-learning has become very famous in recent years. This is the modern era of internet and technology and a lot of information is available on the internet. E-learning helps the learner in accessing the best information available for learning purpose. In this bundle of endless information, not every material is useful for every learner. So, in order to filter the available contents, a process called recommendation is defined which help users in recommending the most appropriate material for learning. One of the main issues with recommender systems is called the new-user cold-start problem. When a new user enters in a system, there is very little information present about the user and this causes inaccuracy in recommendation for the user. In this paper, we have proposed an approach which increases the recommendation accuracy for new-user cold-start problem. The evaluation of the proposed approach shows the improvement by 6% from the existing approach.