Improving the Accuracy of Recommendations and Trust in Collaborative Filtering Approach using LBG Clustering, Levy, and Chaotic Fruit Fly Algorithms

Nowadays, recommender systems play a major role in every aspect of our life. The most commonly used type of recommender systems are collaborative filtering based systems, which have become highly popular. In this paper, a novel approach is proposed for making recommendations in recommender systems. This approach combines collaborative filtering with LBG clustering in order to improve the accuracy of recommendations. In the proposed approach, Levy algorithm is integrated with chaotic fruit fly to tune the parameters and optimize the clusters. The purpose of this approach is to enhance the quality, accuracy, and security through considering the trust parameter using subjective logic model in explicit and implicit forms. The results of experiments conducted on the Movie lens dataset suggest a decrease in error and improvement in accuracy and security comparing to the existing methods.

Keywords:

Recommendation Systems , LBG Clustering , Levy Flight , Chaotic Fruit Fly , Trust