Justifled Recommendations based on Content and Rating Data

Providing justiflcation to a recommendation gives credibil- ity to a recommender system. Some recommender systems (Amazon.com etc.) try to explain their recommendations, in an efiort to regain cus- tomer acceptance and trust. But their explanations are poor and un- justifled, because they are based solely on rating or navigational data, ignoring the content data. In this paper, we propose a novel approach that attains simultaneously accurate and justiflable recommendations. We construct a feature proflle for the users, to reveal their favorite fea- tures. Moreover, we create biclusters (i.e. group of users which exhibit highly correlated ratings on groups of items) to exploit partial matching between the preferences of the test user and each community of users. We have evaluated the quality of our justiflcations with an objective metric in a real data set, showing the superiority of the proposed approach. We also conducted a user study to measure users' satisfaction against the existing and our proposed justiflcation style. The user study shows that our justiflcation style is users' favorite choice.