A Ranking Method for Multimedia Recommenders

In the last few years, recommender systems have gained sig- ni¯cant attention in the research community, due to the in- creasing availability of huge data collections, such as news archives, shopping catalogs, or virtual museums. In this scenario, there is a pressing need for applications to pro- vide users with targeted suggestions to help them navigate this ocean of information. However, no much e®ort has yet been devoted to recommenders in the ¯eld of multimedia databases. In this paper, we propose a novel approach to rec- ommendation in multimedia browsing systems, based on an importance ranking method that strongly resembles the well known PageRank ranking system. We model recommenda- tion as a social choice problem, and propose a method that computes customized recommendations by originally comb- ing intrinsic features of multimedia objects, past behavior of individual users and overall behavior of the entire commu- nity of users. We implemented a prototype of the proposed system and preliminary experiments have shown that our approach is promising.