Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis
Privug provides privacy researchers with a fast and lightweight way to experiment with and evaluate privacy protection measures and mechanisms, addressing the critical need for diligent investigation of privacy risks before data disclosure.
This paper introduces Privug, a tool-supported method that reinterprets data analytics and anonymization programs probabilistically to quantify information leakage. It uses off-the-shelf Bayesian inference tools for information-theoretic analysis, demonstrating accuracy, scalability, and applicability across various leakage analysis scenarios.
Disclosure of data analytics results has important scientific and commercial justifications. However, no data shall be disclosed without a diligent investigation of risks for privacy of subjects. Privug is a tool-supported method to explore information leakage properties of data analytics and anonymization programs. In Privug, we reinterpret a program probabilistically, using off-the-shelf tools for Bayesian inference to perform information-theoretic analysis of the information flow. For privacy researchers, Privug provides a fast, lightweight way to experiment with privacy protection measures and mechanisms. We show that Privug is accurate, scalable, and applicable to a range of leakage analysis scenarios.