王国军2,,
1.中南大学信息科学与工程学院 ??长沙 ??410083
2.广州大学计算机科学与教育软件学院 ??广州 ??510006
基金项目:国家自然科学基金(61632009, 61472451)，广东省自然科学基金(2017A030308006)，广东省高等教育高层次人才计划(2016ZJ01)，中南大学中央高校基本科研业务费专项资金 (2017zzts141)
详细信息
作者简介:张强：男，1988年生，博士生，研究方向为隐私保护、个性化搜索
王国军：男，1970年生，教授，博士生导师，研究方向为云计算、大数据、隐私保护
通讯作者:王国军 csgjwang@gmail.com
中图分类号:TP393计量
文章访问数:1201
HTML全文浏览量:330
PDF下载量:19
被引次数:0
出版历程
收稿日期:2017-12-04
修回日期:2018-04-20
网络出版日期:2018-06-07
刊出日期:2018-08-01
Privacy Preserving Method Based on Location Service in Personalized Search
Qiang ZHANG1,Guojun WANG2,,
1. School of Information Science and Engineering, Central South University, Changsha 410083, China
2. School of Computer Science and Educational Software, Guangzhou University, Guangzhou 510006, China
Funds:The National Natural Science Foundation of China (61632009, 61472451), The Guangdong Provincial Natural Science Foundation (2017A030308006), The High-Level Talents Program of Higher Education in Guangdong Province (2016ZJ01), The Fundamental Research Funds for the Central Universities of Central South University (2017zzts141)
摘要
摘要:在基于位置服务的个性化搜索中，利用可信第三方服务器以及对等节点是保护用户隐私的主要方法，但在现实生活中，它们却是不完全可信的。为了解决这一问题，该文提出一种个性化搜索中基于位置服务的隐私保护方法。该方法通过转换用户的位置信息，并根据用户的查询类型生成用户模型，进而形成带有用户位置信息的查询矩阵，然后利用矩阵加密用户的查询，隐藏查询矩阵中的用户信息，最后根据安全内积计算返回相关性得分最高的前K个查询文件给用户。安全性分析表明该方法能有效地保护用户的查询隐私和位置隐私，通过分析与实验表明，该方法大幅度地缩短了索引构建时间，降低了通信开销，同时为用户提供了基于位置的个性化搜索结果，一定程度上解决了移动设备屏幕小带来的弊端。
关键词:隐私保护/
个性化搜索/
位置转换/
安全内积计算
Abstract:For personalized search based on location service, the trusted third-party server and peer node are used as the main method for privacy preserving. However, entirely trusted third-party server or peer node does not exist in real life. In order to address this problem, a method of privacy preserving on the location of mobile users is proposed when using personalized search. The method is used to convert the user’s location information into distance information and generate the user model according to the user’s query type, forming a query matrix with user location information, then the matrix is used to encrypt the user’s query and conceal the user information in the query matrix. Finally, according to the calculation of the security inner product, the K file with the highest relevance score is returned to the user. It is evident from the security analysis that the proposed method can effectively protect the user’s query privacy and location privacy. The analysis and experimental results show that the proposed method can greatly shorten the time of index construction and reduce the communication overhead. While providing users with location based personalized search results, the method is able to remedy the defects of small-screen mobile devices.
Key words:Privacy preserving/
Personalized search/
Location conversion/
Security inner product
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=0fc3f650-51cd-47b4-8224-5489f59ccfc5
