Impact of Relevance Measures on the Robustness and Accuracy of Collaborative Filtering

Abstract

The open nature of collaborative recommender systems present a security problem. Attackers that cannot be readily distinguished from ordinary users may inject biased profiles, degrading the objectivity and accuracy of the system over time. The standard user-based collaborative filtering algorithm has been shown quite vulnerable to such attacks. In this paper, we examine relevance measures that complement neighbor similarity and their influence on algorithm robustness. In particular, we consider two techniques, significance weighting and trust weighting, that attempt to calculate the utility of a neighbor with respect to rating prediction. Such techniques have been used to improve prediction accuracy in collaborative filtering. We show that significance weighting, in particular, also results in improved robustness under profile injection attacks.