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.
作者
齐晓娜
王佳
徐东升
张宇敬
郭佳
刘阳
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)
出处
《河北大学学报（自然科学版）》
CAS
北大核心
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)
