BROWSE
Detail
Conference Paper
Advanced Trajectory Privacy Protection with Attention Mechanism and Auxiliary Classifier Generative Adversarial Networks
Cited 1 time in 
Share 
Share
- Authors
- Jihwan Shin, Yeji Song, Yoo-Young Cheong, Jinhyun Ahn, Taewhi Lee, Dong-Hyuk Im
- Issue Date
- 2024-01
- Citation
- International Conference on Information Networking (ICOIN) 2024, pp.257-261
- Language
- English
- Type
- Conference Paper
- Abstract
- Advancements in smartphone technology have led to an increase in the usage of various location-based services (LBS). This resulted in a lot of trajectory data being generated. LBSs provide personalized services for users through continuous queries. However, there is a problem that the user's sensitive information can be inferred through such continuous queries. Although various methods have been proposed to protect personal information, traditional personal information protection methods cannot provide absolute personal information protection when user location information itself, such as LBS, is required. In particular, if sensitive points visited by the user are exposed, this may lead to additional information leakage. Therefore, this paper guarantees anonymity by protecting sensitive points visited by users through the class conditional synthesis of ACGAN, and proposes a synthetic trajectory generation model for generating highly useful trajectory data through the combination of attention mechanisms. Furthermore, the usability and anonymity aspects of the synthetic trajectory data generated by the proposed are compared with those of existing models to verify the performance of the proposed.
- KSP Keywords
- Attention mechanism, Continuous queries, Generation model, Location information, Location-Based Services, Personalized service, Protection method, Sensitive information, Trajectory Data, Trajectory Privacy Protection, additional information