Privacy-Preserving Distributed Collaborative Filtering

We propose a mechanism to preserve privacy while leveraging user profiles in distributed recommender systems. Our approach relies on (i ) an original obfuscation mechanism hiding the exact profiles of users without significantly decreasing their utility, as well as (ii ) a randomized dissemination algorithm ensuring differential privacy during the dissemination process. We evaluate our system against an alternative providing differential privacy both during profile construction and dissemination. Results show that our solution preserves accuracy without the need for users to reveal their preferences. Our approach is also flexible and more robust to censorship