Recommending with an Agenda: Active Learning of Private Attributes using Matrix Factorization

TL;DR

This paper demonstrates how recommender systems can infer private user attributes through active learning attacks using Bayesian matrix factorization, raising privacy concerns without degrading recommendation quality.

Contribution

It introduces novel passive and active attack methods leveraging Bayesian matrix factorization to infer private attributes efficiently and stealthily.

Findings

Attacks require fewer ratings than static methods

Attacks do not reduce recommendation quality

Feasibility demonstrated on multiple datasets

Abstract

Recommender systems leverage user demographic information, such as age, gender, etc., to personalize recommendations and better place their targeted ads. Oftentimes, users do not volunteer this information due to privacy concerns, or due to a lack of initiative in filling out their online profiles. We illustrate a new threat in which a recommender learns private attributes of users who do not voluntarily disclose them. We design both passive and active attacks that solicit ratings for strategically selected items, and could thus be used by a recommender system to pursue this hidden agenda. Our methods are based on a novel usage of Bayesian matrix factorization in an active learning setting. Evaluations on multiple datasets illustrate that such attacks are indeed feasible and use significantly fewer rated items than static inference methods. Importantly, they succeed without sacrificing the quality of recommendations to users.