基于改进的k-means差分隐私保护方法在位置隐私保护中的应用  

Application of improved k-means differential privacy protection in location privacy protection

作　　者：齐晓娜[1] 王佳 徐东升[3] 张宇敬[1] 郭佳[1] 刘阳[1] QI XiaonaI;WANG Jia;XU Dongsheng;ZHANG Yujing;GUO Jia;LIU Yang(Department of Information Management ＆Engineering,Hebei Finance University,Baoding071051,China;Experimental Teaching Center,Hebei Finance University,Baoding 071051,China;Department of Computer Applied Engineering,Hebei Software Institute,Baoding 071000,China)

机构地区：[1]河北金融学院信息管理与工程系,河北保定071051 [2]河北金融学院实验教学中心,河北保定071051 [3]河北软件职业技术学院计算机应用工程系,河北保定071000

出　　处：《河北大学学报（自然科学版）》2018年第3期315-320,共6页Journal of Hebei University(Natural Science Edition)

基　　金：河北省高等学校科学技术研究项目(QN2017327)

摘　　要：针对k-means差分隐私聚类结果的可用性较差的问题,依据LBS的数据采集特点对k-means算法进行了改进.仿真实验证明:在LBS隐私保护方面,提出的改进k-means聚类方法在聚类结果的匿名性方面相对普通差分隐私k-means聚类方法有一定程度的提高.In view of the poor availability of k-means differential privacy clustering results,the kmeans algorithm is improved on the basis of the characteristics of the data acquisition of LBS.Proved by simulation experiment,The newk-means clustering method proposed in this paper had a certain degree of improvement in the anonymity of clustering results than the ordinary differential privacy k-means clustering method in terms of LBS privacy protection.

关 键 词：K-MEANS 聚类 差分隐私 位置隐私保护 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]