Privacy Protection Of User Relations Based On Trajectory Data

Various location-based service applications have brought great convenience to people's lives.At the same time,users' privacy issues have received more and more attention.The trajectory data contains a variety of information about the user,and the attacker can discover the privacy of the user's relationship from the trajectory.The goal of this paper is to protect the privacy of user re-lationships.The solution proposed in this paper is universal and can be applied to software related to user relationships,such as Weibo and WeChat and so on.In this paper,an algorithm for discovering user intimacy relationship is proposed in the context of trajectory data,and a corresponding protection algorithm is proposed for intimate relationships that do not meet privacy requirements.(1)This article first discusses how to measure the intimacy among users and proposes an algorithm for calculating user intimacy(CUIA).Based on the similar sub-trajectories,the algorithm uses the factors designed to measure the user's intimacy relationship to calculate the intimacy among users.The factors include time similarity,location similarity and sequence similarity,including sensitive time and sensitive location,and finally normalized by Sigmoid function.Using formulas to derive the degree of intimacy among users.The algorithm can accurately measure the intimacy between users.(2)This paper proposes a user relationship privacy protection model,kmn-anonymity.This anonymous model can reach the user's privacy requirements faster.Based on the anonymous model,the user relationship that does not meet the anonymous requirement is found from the relationship diagram,and divides the user relationships that need to be protected into three categories in the graph structure,namely:single intimate relationship,transit relationship and multi-transfer relationship.(3)This paper proposes algorithms to protect user intimacy,namely global generalization algorithm(GGA)and local generalization algorithm(LGA).The global generalization algorithm can effectively protect the user's privacy,but it also makes the data distortion more serious,which makes the data availability lower.To solve this problem,this paper proposes a local generalization method,according to the above three different user relationships.The relationship is designed with different protection algorithms for subtracting relationships,adding relationships,and adding false data.The technical means used include:time generalization,location generalization,time disturbance,location disturbance and adding false data.In addition,this paper proposes the relationship between the two for the anonymous model of personal protection and relationship protection.(4)Finally,the experimental environment is built for the proposed algorithm,and the measurement index of the algorithm is proposed.The algorithm and other algorithms proposed in this paper are compared and analyzed from the aspects of data availability.The experimental results prove that the algorithm in this paper is effective.