Two Differentially Private Rating Collection Mechanisms for Recommender Systems

04/28/2016

∙

We design two mechanisms for the recommender system to collect user ratings. One is modified Laplace mechanism, and the other is randomized response mechanism. We prove that they are both differentially private and preserve the data utility.

READ FULL TEXT

Two Differentially Private Rating Collection Mechanisms for Recommender Systems

Exponential Randomized Response: Boosting Utility in Differentially Private Selection

Colored Noise Mechanism for Differentially Private Clustering

Differentially Private Functional Summaries via the Independent Component Laplace Process

Differentially Private Sketches for Jaccard Similarity Estimation

Differentially Private Learning of Undirected Graphical Models using Collective Graphical Models

Test without Trust: Optimal Locally Private Distribution Testing

Observations on the Bias of Nonnegative Mechanisms for Differential Privacy

Two Differentially Private Rating Collection Mechanisms for Recommender Systems

Related Research

Exponential Randomized Response: Boosting Utility in Differentially Private Selection

Colored Noise Mechanism for Differentially Private Clustering

Differentially Private Functional Summaries via the Independent Component Laplace Process

Differentially Private Sketches for Jaccard Similarity Estimation

Differentially Private Learning of Undirected Graphical Models using Collective Graphical Models

Test without Trust: Optimal Locally Private Distribution Testing

Observations on the Bias of Nonnegative Mechanisms for Differential Privacy