On the Evaluation of the Privacy Breach in Disassociated Set-Valued Datasets

TL;DR

This paper investigates the privacy risks of disassociation, a bucketization technique for anonymizing set-valued datasets, revealing potential privacy breaches and evaluating their severity with a quantitative detection method.

Contribution

It demonstrates the limitations of disassociation in preventing privacy breaches and evaluates these risks using real datasets and a novel detection algorithm.

Findings

Privacy breaches can occur due to cover problems in disassociated datasets.

Disassociation's limits are demonstrated through real dataset analysis.

Quantitative methods can detect privacy breaches effectively.

Abstract

Data anonymization is gaining much attention these days as it provides the fundamental requirements to safely outsource datasets containing identifying information. While some techniques add noise to protect privacy others use generalization to hide the link between sensitive and non-sensitive information or separate the dataset into clusters to gain more utility. In the latter, often referred to as bucketization, data values are kept intact, only the link is hidden to maximize the utility. In this paper, we showcase the limits of disassociation, a bucketization technique that divides a set-valued dataset into $k^m$-anonymous clusters. We demonstrate that a privacy breach might occur if the disassociated dataset is subject to a cover problem. We finally evaluate the privacy breach using the quantitative privacy breach detection algorithm on real disassociated datasets.