A trajectory sensitive data protection method supporting personalized privacy

Abstract

The invention provides a trajectory sensitive data protection method supporting personalized privacy, which relates to the technical field of data analysis and processing. The method comprises the following steps of: extracting a sensitive position list SList which is set by a user in an original data set D and needs to be protected; using the Google Map API to reverse address parse any sensitiveposition SPb in the user-set sensitive position list SList, and obtaining the sensitive sequence SSb; constructing false trajectory fragments according to the cluster set Clus; replacing a trajectorysegment T in that original data set D with a false trajectory segment T*, and outputting a privacy preserved data set D*. The trajectory sensitive data protection method supporting personalized privacy considers the moving speed of the moving object, the number of sampling points and the similarity of the road sections, the sensitive positions are divided into semantic clusters, and the substitution of the sensitive positions is random. The probability of the attacker to determine the true sensitive positions is much lower, which ensures the security of the sensitive positions.

Description

technical field [0001] The invention relates to the technical field of data analysis and processing, in particular to a track-sensitive data protection method supporting personalized privacy. Background technique [0002] In recent years, with the rapid development of positioning technology, mobile devices with GPS receivers have become more and more popular, and location-based services (LBS) have become more and more widely used, so service providers will obtain a large amount of trajectory data every day . Data mining and analysis of collected trajectory data are widely used in many fields, such as urban planning, business decision-making, travel route recommendation, etc. However, trajectory data contains personal sensitive information. By analyzing the released trajectory data, attackers may cause privacy leaks such as personal hobbies and behavior patterns. At present, for the trajectory privacy protection problem in data release, most of them are based on the traditi...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the problems existing in the prior art, the present invention provides a track-sensitive data protection method that supports personalized privacy, considers different privacy requirements of different users, and solves the leakage of sensitive positions in user tracks during data publishing. Visit probability, number of user visits and average stay time of users in different time periods, consider the spatial distance and semantic similarity of sensitive locations, determine the mixing distance between two points, and then cluster and divide sensitive locations based on the principle of l-diversity. Sensitive positions are replaced, and a false trajectory segment is constructed according to the moving speed of the moving object, the number of sampling points and the sampling interval, so as to improve the availability of published data

Images