Item-Based Energy Clustering Recommendation

Previous recommendation systems have focused on algorithms to make the recommendations based on the individual items. However, in many areas, the introduction about a cluster of the items based on the general characteristics of the item is more important than just focusing on the individual items. In this paper, we have proposed a new approach for the recommendation system, the proposed method uses the energy distance to group the items with similar properties or characteristics into a cluster, then based on the item clusters to give the most suitable recommendations for the users. In addition, the methods based on error (MAE_(c)) and accuracy (Precison_(c)-Recall_(c)) are also selected to evaluate the reliability of the new proposed model on two popular datasets Jester5k and MovieLens100k. Besides, the proposed model is also compared with two item-based collaborative filtering models using the Cosine and Pearson measures in “rrecsys” package and three item-based collaborative filtering models using the Matching, Euclidean and Karypis measures in “recommenderlab” package. The experimental results have shown that the proposed model is better than the compared models.