Research On Privacy Preserving High-dimensional Trajectory Data Publishing Method Based On Suppression Technology

In recent years,with the rapid development of mobile communication technology and GPS technology,service providers and research institutes accumulate a large number of users' location data containing abundant information,which promotes the development of the trajectory data mining research.However,trajectory data usually contains a lot of sensitive information,which is subject to leak while the trajectory data is published for potential information mining,User privacy protection has been paid more and more attention by researchers both at home and abroad.Therefore,how to protect user privacy in the process of trajectory data publishing has become an important issue in data privacy protection.This thesis mainly focuses on existing user privacy problem in high-dimensional trajectory data publishing process,analyzes the existing privacy protection technology,and proposes our privacy preserving algorithm for the trajectory sequence causes user privacy disclosure and trajectory combined with non-sensitive information to cause user privacy disclosure respectively.The main work is as follows:1)In order to resist the identity linkage attack and attribute linkage attack caused by the trajectory sequence in the high-dimensional trajectory publishing,and consider the mining usability of the dataset after anonymous processing,a high-dimensional trajectory data privacy-preserving algorithm based on Maximal Frequent Sequence(MFS)analysis was proposed.At First,the algorithm uses the existence rule of MFS to design the decision model of the spatio-temporal point suppression order.Then,to ensure user privacy security that using the suppression operation to eliminate the trajectory sequence existing privacy reveal,to reduce the data loss of usability that using the reconstruction operation to write the local suppression spatio-temporal point back to the dataset.At last,the proposed algorithm is compared with other existing algorithms from data utility,and the result of experiments show that the effectiveness of the proposed algorithm.2)Aiming at the issue of privacy threats between high-dimensional trajectory data and non-sensitive information,a high-dimensional trajectory data privacy preserving algorithm based on non-sensitive information analysis was proposed.Firstly,the algorithm analyzed the correlation between trajectory and non-sensitive information to build privacy disclosure decision model.Secondly,by common sequence,the spatio-temporal point,which caused the minimal loss of information,was selected as the suppression member while eliminate the risk of privacy disclosure associated with the combination of trajectory and non-sensitive information.Finally,the anonymized trajectory dataset that preserves privacy and low data loss was generated.The experiment results show that the proposed algorithm can ensure the trajectory k-anonymity,and reduce the usability loss of data.The proposed algorithm can be effectively solved to the privacy reveal problem of high-dimensional trajectory data publishing.