Research On Protecting Spatio-Temporal Pattern Privacy In Social Networks

In recent years,with the rapid development of social networks and the continuous improvement of positioning technologies,spatio-temporal data in social network data are exploding.There is a risk of privacy leakage during the spatio-temporal data publishing.Therefore,data privacy protection for social networks has become a research hotspot.Generaly,a user in social networks follows a certain periodic law to check in at the same location within a fixed time period,so that all the published spatio-temporal data have periodic laws.However,the existing privacy protection mechanism does not consider such spatiotemporal laws,so that the published spatio-temporal data may reveal users' behavioral pattern privacy with spatio-temporal laws.To solve this problem,we define and study the technology of protecting the spatio-temporal pattern privacy in social networks for the first time.Firstly,in this paper,we elaborate on the mechanism for inferring the spatio-temporal pattern privacy;we propose to use the technology of combining suppression with generalization to protect the spatio-temporal pattern privacy;we define the related information loss function to evaluate the information loss caused by the privacy protection.These lay the foundation for the design and implementation of protecting the spatio-temporal pattern privacy.Secondly,a Spatio-Temporal Pattern Privacy Protection algorithm(denoted as STPPP)is proposed.In STPPP,we suppress and generalize certain spatio-temporal data to protect spatiotemporal pattern privacy;we design heuristic rules to preferentialy tackle the spatio-temporal data with high protection contribution and low information loss,which ensures high availability of the published spatio-temporal data while protecting the spatio-temporal pattern privacy;spatio-temporal data are generalized based on users' mobile patterns and spatio-temporal accessibility between locations,which makes the published data be more in line with the users' behavior patterns and prevents attackers from identifying the real data;a strategy of updating data is proposed to optimize the executing efficiency of the STPPP algorithm.Finally,extensive experiments on real datasets and comparisons with the existing methods demonstrate the Spatio-Temporal Pattern Privacy Protection algorithm can effectively protect the spatio-temporal pattern privacy while ensuring high executing efficiency of the STPPP algorithm and high availability of the published spatio-temporal data.