Position representative element generation method for track data release

Abstract

A position representative element generation method for track data release specifically comprises the following steps: 1) based on position point index construction of a Hilbert index value curve, generating position clusters at all moments; 2) according to the privacy budget configuration of each position cluster, generating a sampling position cluster of each position cluster by using a samplingmechanism based on a privacy budget threshold; 3) applying a differential privacy index mechanism to each sampling position cluster to select a representative element of the position cluster; and 4)replacing the original position point in each position cluster with the position representative element to realize generalization of the original trajectory data. The method provided by the inventioncan ensure that the privacy protection requirement of a data provider is satisfied in a trajectory data publishing mode, provides strict privacy guarantee, and improves the availability of the trajectory published data.

Description

technical field [0001] The invention belongs to the technical field of database processing, and in particular relates to a method for generating position representative elements used for track data release. Background technique [0002] With the widespread application of location-based services and smart devices, a huge amount of spatial data, especially user trajectory data, has been generated. Through the mining and analysis of trajectory data, the basic design of the city can be effectively designed, thereby effectively avoiding resource waste and traffic congestion. However, trajectory data usually contains some personal sensitive information, and simple processing methods may lead to serious privacy leakage problems. Therefore, meeting the privacy protection needs of data providers is a key issue to be solved in trajectory data release. [0003] As a privacy protection technology with a strict theoretical basis, differential privacy has been widely used in trajectory ...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the deficiencies of the above-mentioned prior art, the object of the present invention is to provide a method for generating position representative elements for trajectory data distribution, which solves the problems of low privacy protection effect and data availability in the distribution of trajectory data in the prior art , by clustering the location point sets at different times, the location clusters and the corresponding privacy budget configurations are obtained, and then the sampling mechanism based on the privacy budget threshold is used to extract the sampling location clusters, and the index of differential privacy is applied to the sampling location clusters The mechanism selects the representative unit of the cluster at this location, which meets the individual privacy requirements of the data provider, provides strict privacy guarantees, and improves the availability of trajectory release data

Images