BLC: Private Matrix Factorization Recommenders via Automatic Group Learning

TL;DR

This paper introduces BLC, a privacy-preserving matrix factorization method that leverages user grouping to enhance privacy without sacrificing recommendation accuracy.

Contribution

It presents a novel group-based matrix factorization approach and the BLC algorithm for privacy-enhanced recommendations in shared user groups.

Findings

Privacy is improved through user grouping

Recommendation accuracy remains high

BLC outperforms existing privacy-preserving methods

Abstract

We propose a privacy-enhanced matrix factorization recommender that exploits the fact that users can often be grouped together by interest. This allows a form of "hiding in the crowd" privacy. We introduce a novel matrix factorization approach suited to making recommendations in a shared group (or nym) setting and the BLC algorithm for carrying out this matrix factorization in a privacy-enhanced manner. We demonstrate that the increased privacy does not come at the cost of reduced recommendation accuracy.