PrivSem : Protecting location privacy using semantic and differential privacy

In this paper, we address the problem of users’ location privacy preservation on road networks. Most existing privacy preservation techniques rely on structure-based spatial cloaking , but pay little attention to locations’ semantic information. Yet, the semantics may disclose sensitive information of mobile users. In addition, these studies ignore the location privacy requirements of other users, which is essential for location-based services (LBS). Thus, to tackle these problems, we propose PrivSem , a novel framework which integrates location k - anonymity , segment l - semantic diversity , and differential privacy to protect user location privacy from violation. In this framework, rather than using the original location data, we only access to the sanitized data according to differential privacy. Due to the nature of differential privacy which perturbs the real data with noise, it is particularly challenging to determine an effective cloaked area . Further, we investigate an error analysis model to ensure the effectiveness of the generated cloaked areas. Finally, through formal privacy analysis, we show that our proposed approach is effective in providing privacy guarantees. Extensive experimental evaluations on large real-world datasets are conducted to demonstrate the efficiency and effectiveness of PrivSem .