PCTA: privacy-constrained clustering-based transaction data anonymization

Abstract

Transaction data about individuals are increasingly collected to support a plethora of applications, spanning from marketing to biomedical studies. Publishing these data is required by many organizations, but may result in privacy breaches, if an attacker exploits potentially identifying information to link individuals to their records in the published data. Algorithms that prevent this threat by transforming transaction data prior to their release have been proposed recently, but incur significant information loss due to their inability to accommodate a range of different privacy requirements that data owners often have. To address this issue, we propose a novel clustering-based framework to anonymizing transaction data. Our framework provides the basis for designing algorithms that explore a larger solution space than existing methods, which allows publishing data with less information loss, and can satisfy a wide range of privacy requirements. Based on this framework, we develop PCTA, a generalization-based algorithm to construct anonymizations that incur a small amount of information loss under many different privacy requirements. Experiments with benchmark datasets verify that PCTA significantly outperforms the current state-of-the-art algorithms in terms of data utility, while being comparable in terms of efficiency.