Collaborative filtering over distributed environment

Currently, implementations of the Collaborative Filtering (CF) algo- rithm are mostly centralized. Hence, information about the users, for example, product ratings, is concentrated in a single location. In this work we propose a novel approach to overcome the inherent limitations of CF (sparsity of data and cold start) by exploiting multiple distributed information repositories. These may belong to a single domain or to different domains. To facilitate our ap- proach, we used LoudVoice, a multi-agent communication infrastructure that can connect similar information repositories into a single virtual structure called "implicit organization". Repositories are partitioned between such or- ganizations according to geographical or topical criteria. We employ CF to generate user-personalized recommendations over different data distribution policies. Experimental results demonstrate that topical distribution outperforms geographical distribution. We also show that in geographical distribution using filtering based on social characteristics of the users improves the quality of rec- ommendations.