Privacy-preserving Collaborative Filtering Using Randomized Response

Abstract

This paper proposes a new privacy-preserving recommendation method classified into a randomized perturbation scheme in which a user adds a random noise to the original rating value and a server provides a disguised data to allow users to predict the rating value for unseen items. The proposed scheme performs a perturbation in a randomized response scheme, which preserves a higher degree of privacy than that of an additive perturbation. To address the accuracy reduction of the randomized response, the proposed scheme uses a posterior probability distribution function, derived from Bayes' estimation for the reconstruction of the original distribution, to revise the similarity between items computed from the disguised matrix. A simple experiment shows the accuracy improvement of the proposed scheme.