Community based recommendation in e-learning systems

Abstract

The success of any e-learning system depends on quality and the quantity of assistance provided to its students, in the learning process. Hence, it is essential to analyze a student’s academic skills in order to personalize the education provided both vertically and horizontally. This paper proposes a novel approach through which initially students are grouped based on several factors including their academic interests and further motivate the students to enhance their knowledge by providing appropriate recommendations made based on students belonging to their group. It has been proved that neither link information nor content information individually is sufficient to form student communities (Rabbany et al., 2011). Therefore, the approach includes both together for community detection. Further, the approach also intends to recommend courses based on the ratings for courses given by other students with similar skill sets in the same group. Experimental results highlight the quality or relevance of the recommendations made within communities, which in turn reflects on the accuracy of the proposed community detection method.