Experience-based Personalized Diversification of Recommendations

Abstract

Recommender Systems are fundamentally directed towards offering a worthwhile assistance to users, to choose something useful from an overwhelming set of options. Although accuracy of the recommendations has been regarded as the primary quality aspect of recommender systems, there’s an increasing cognizance that there are other factors such as diversity that users also value. Despite the increased interest of researchers to improve diversification of recommendations, we find that personalization of diversification has been overlooked. This work offers two main contributions. PersonalizedDiv technique which diversifies recommendation lists based on the user’s past behavior. Personalization is achieved by diversifying the recommendation list with more novel items if the user has shown diverse preferences in the past, and diversifying the recommendation list with more relevant items if the user has shown homogeneous preferences in the past. The proposed technique is capable of controlling the tradeoff between accuracy and diversity while consolidating personalization. Moreover, as obtaining a diversified recommendation list by conventional recommender systems is unlikely, current approaches generally obtain diverse results by selecting items from a larger recommendation list generated by some recommender system given a diversification technique. Thus, we find that the recommender technique used to generate the initial recommendation list is also important to the diversification procedure. Consequently, we propose ExpertRec technique which uses the ratings of users and also the ratings of experienced item category experts in recommendation generation process in order to generate a better initial recommendation list with novel and relevant items to improve the diversification process. Our experiments and evaluation provides evidence to illustrate the properties of proposed techniques and indicate the proposed approach has comparable results to state-of-art techniques. Moreover, unlike other techniques, our approach can promote both novel and relevant items and also make the diversification process personalized.