Robustness Analyses of Instance-Based Collaborative Recommendation

Collaborative recommendation has emerged as an effective technique for a personalized information access. However, there has been relatively little theoretical analysis of the conditions under which the technique is effective. We analyze the robustness of collaborative recommendation: the ability to make recommendations despite (possibly intentional) noisy product ratings. We formalize robustness in machine learning terms, develop two theoretically justified models of robustness, and evaluate the models on real-world data. Our investigation is both practically relevant for enterprises wondering whether collaborative recommendation leaves their marketing operations open to attack, and theoretically interesting for the light it sheds on a comprehensive theory of collaborative recommendation.