On Privacy of Socially Contagious Attributes

TL;DR

This paper investigates how social contagion influences the privacy guarantees of differential privacy mechanisms, showing that knowledge of social networks can compromise privacy and proposing methods to address this challenge.

Contribution

It provides a theoretical analysis of differential privacy under social contagion, revealing limitations and proposing improved privacy-preserving techniques considering social influence.

Findings

Adversaries cannot surpass a specific AUC threshold without network knowledge.

Knowledge of contagion networks enables better prediction of sensitive data.

Nodes with high social influence are more vulnerable to privacy breaches.

Abstract

A commonly used method to protect user privacy in data collection is to perform randomized perturbation on user's real data before collection so that aggregated statistics can still be inferred without endangering secrets held by individuals. In this paper, we take a closer look at the validity of Differential Privacy guarantees, when the sensitive attributes are subject to social influence and contagions. We first show that in the absence of any knowledge about the contagion network, an adversary that tries to predict the real values from perturbed ones, cannot achieve an area under the ROC curve (AUC) above $1-(1-\delta)/(1+e^\varepsilon)$, if the dataset is perturbed using an $(\varepsilon,\delta)$-differentially private mechanism. Then, we show that with the knowledge of the contagion network and model, one can do significantly better. We demonstrate that our method passes the performance limit imposed by differential privacy. Our experiments also reveal that nodes with high influence on others are at more risk of revealing their secrets than others. The performance is shown through extensive experiments on synthetic and real-world networks.