Research on Government Data Publishing Based on Differential Privacy Model

Abstract

With the enforcement of the policies of opening and sharing information resources, protection of citizens' privacy has become a key issue concerned by the government and public. This paper discusses the risk of citizens' privacy disclosure related to government data publishing, and analyzes the main privacy-preserving methods for data publishing. Aiming at the problem that most of the existing privacy protection models for data publishing cannot resist the attacks based on the growing background knowledge, a differential privacy framework for publishing governmental statistical data is established. Based on the framework, a data publishing algorithm using MaxDiff histogram is proposed. Applying differential method, Laplace noises are added to the original dataset, which prevents citizens' privacy from disclosure even if attackers get strong background knowledge. According to the maximum frequency difference, the adjacent data bins are grouped, then the differential privacy histogram with minimum average error can be constructed. Through theoretical analysis and experimental comparison, it is demonstrated that the proposed data publishing algorithm can not only be used to effectively protect citizens' privacy, but also reduce the query sensitivity and improve the utility of the data published.