Community Detection-Based Recommendation Framework

Abstract

Recommender system is a type of information filtering system predicting users preferences about items, aiming to generate personalized recommendations. Various recommendation approaches exist in the literature that differ in terms of methodology and types of systems they can be utilized on. In recent years some attempts have been made to incorporate community detection methods into recommender systems to make the process of recommendation generation more accurate in terms of rating or preference prediction and efficient in terms of computational complexity. In this paper we propose a community detection-based approach for recommender system, which is more reasonable in certain applications.