A Generalization-Based Approach for Personalized Privacy Preservation in Trajectory Data Publishing

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 that are associated with trajectory data. Therefore, improper publishing of the trajectory database could 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 the adversary employs some background knowledge. The existing approaches for privacy preservation in trajectory data publishing apply the same amount of privacy preservation for all moving objects, without regard to their privacy requirements. The consequence is that some moving objects may be offered insufficient privacy preservation, while some others may not need high privacy protection. In this paper, we address this issue and present a novel approach for privacy preservation in trajectory data publishing based on the concept of personalized privacy. It consists of two main steps: (1) identifying primary critical trajectory data records and generalizing sensitive attributes according to them, and (2) identifying remaining critical trajectory data records and eliminating moving points with minimum information loss. The results of experiments on a trajectory dataset show that our proposed approach achieve the conflicting goals of data utility and data privacy in accordance with the privacy requirements of moving objects.