TrPLS: Preserving Privacy in Trajectory Data Publishing by Personalized Local Suppression

Abstract

Trajectory data are becoming more popular due to the rapid development of mobile devices and the widespread use of location-based services. They often provide useful information that can be used for data mining tasks. However, a trajectory database may contain sensitive attributes, such as disease, job, and salary, which are associated with trajectory data. Hence, improper publishing of the trajectory database can put the privacy of moving objects at risk. Removing identifiers from the trajectory database before the public release, is not effective against privacy attacks, especially, when an adversary uses some partial trajectory information as its background knowledge. The existing approaches for preserving privacy in trajectory data publishing apply the same amount of privacy protection for all moving objects without considering their privacy requirements. The consequence is that some moving objects with high privacy requirements may be offered low privacy protection, and vice versa. In this paper, we address this challenge and present TrPLS, a novel approach for preserving privacy in trajectory data publishing. It combines local suppression with the concept of personalization to achieve the conflicting goals of data utility and data privacy in accordance with the privacy requirements of moving objects. The results of experiments on a trajectory dataset show that TrPLS can be successfully used for preserving personalized privacy in trajectory data publishing.

Keywords

• trajectory data
• privacy preservation
• personalized privacy
• quasi-identifier
• local suppression
• information loss
• disclosure risk