Privacy Leakage over Dependent Attributes in One-Sided Differential Privacy

TL;DR

This paper extends One-Sided Differential Privacy to account for dependencies between records, quantifying privacy leakage and optimizing utility-privacy trade-offs in dependent data scenarios.

Contribution

It introduces a method to quantify privacy leakage with dependent records and proposes an optimization framework for utility-privacy trade-offs in OSDP.

Findings

Quantifies privacy leakage with dependent attributes.

Develops an optimization approach for utility-privacy balance.

Extends OSDP framework to real-world dependent data.

Abstract

Providing a provable privacy guarantees while maintaining the utility of data is a challenging task in many real-world applications. Recently, a new framework called One-Sided Differential Privacy (OSDP) was introduced that extends existing differential privacy approaches. OSDP increases the utility of the data by taking advantage of the fact that not all records are sensitive. However, the previous work assumed that all records are statistically independent from each other. Motivated by occupancy data in building management systems, this paper extends the existing one-sided differential privacy framework. In this paper, we quantify the overall privacy leakage when the adversary is given dependency information between the records. In addition, we show how an optimization problem can be constructed that efficiently trades off between the utility and privacy.