Abstract
It is a new attack model to mine user’s activity rule from user’s massive data. In order to solve the privacy leakage problem caused by user tendency in current privacy preserving methods, an extended differential privacy preserving method based on user’s tendency is proposed in the paper. By constructing a Markov chain, and using the Markov decision process, it equivalently expresses user’s tendency as measurable state transition probability, which can transform qualitative descriptions of user’s tendency into a quantitative representation to achieve an accurate measurement of the user tendency. An extended (P,ε)-differential privacy protection method is proposed in the work, by introducing a privacy model parameter R, it combines the quantified user’s propensity probability with a differential privacy budget parameter, and it can dynamically add different noise amounts according to the user’s tendency, so as to achieve the purpose of protecting the user’s propensity privacy information and improve data availability. Finally, the feasibility and effectiveness of the proposed method was verified by experiments.
Funder
Changzhou University Doctoral Research Funding Project
Jilin Province Science and Technology Research Planning Project
Publisher
Public Library of Science (PLoS)
Reference42 articles.
1. A differential privacy preserving deep learning caching framework for heterogeneous communication network systems;Wang Huanhuan;International Journal of Intelligent Systems,2022
2. Xiong. An adaptive federated learning scheme with differential privacy preserving[J];Wu Xiang;Future Generation Computer Systems,2022
3. Sharing without caring: privacy protection of users’ spatio-temporal data without compromise on utility;Ozeki Ren;Proceedings of the 30th International Conference on Advances in Geographic Information Systems,2022
4. A Comprehensive Survey on Local Differential Privacy;Xiong Xingxing;Security and Communication Networks,2020
5. Utility-aware and Privacy-preserving Trajectory Synthesis Model that Resists Social Relationship Privacy Attacks;Zheng Zhirun;ACM Transactions on Intelligent Systems and Technology,2022