The effectiveness of user-centric social interfaces on evaluating mobile recommendations

Abstract

Many studies on recommender systems have focused on increasing their accuracy by enhancing the algorithms employed. Social perceptions, however, influence both satisfaction and perceived accuracy of mobile recommender systems. Thus, the aim of this study was to investigate the importance of including reasons for particular recommendations by examining the role of social presence and self-reference and their effects on user evaluations of mobile recommender systems. We performed a 2 x 2 experimental setting (Four experimental web pages with user-to-user and item-to-item collaborative filtering) which was used to manipulate customer perception of accuracy through social presence and self-reference. Social presence and self-reference were shown to be antecedents of perceived accuracy of mobile recommender systems. Additionally, perceived accuracy appeared to be a partial mediator of the relationship between social presence and satisfaction, whereas perceived accuracy appeared to be a full mediator of the relationship between self-reference and satisfaction.