隐私保护的数据挖掘方法的研究-《计算机技术与发展》

文章信息/Info

Author(s):: TANG Lin; HE Feng; School of Computer Science and Engineering,North University for Ethnics

Keywords:: privacy-preserving data mining; privacy-preserving association rule mining; privacy-preserving classification mining; cluster mining; centralized data; distributed data

摘要:: 介绍了隐私保护数据挖掘方法的产生背景和意义,其次概括了现阶段国内外隐私保护数据挖掘算法的研究现状,并对当前隐私保护数据挖掘领域中已提出的算法按照数据挖掘的方法、数据源分布情况、隐私保护技术和隐私保护对象以及数据挖掘应用类型等方面进行分类,然后分别详细阐述了在集中式和分布式数据分布环境下,应用在隐私保护的关联规则挖掘、分类和聚类挖掘中的一些典型的技术和算法,总结出它们的优缺点,并对这些优缺点进行剖析和对比,最后指明了隐私保护数据挖掘算法在未来的整体发展方向

Abstract:: Introduce the background and the significance of the privacy-preserving data mining methods.Secondly,summarized the present research status of the privacy-preserving data mining algorithm at home and abroad,and these algorithms in this area has been proposed already are classified according to the data mining methods,original data distribution,privacy-preserving techniques,privacy protection and data mining application type.Thirdly,it elaborates some typical technologies and algorithms which are used to the type such as the privacy-preserving association rule mining,and the privacy-preserving classification mining,and also the privacy-preserving cluster mining on the environment of the centralized data and the distributed data respectively.Most importantly,it also summarizes their advantages and disadvantages and then analyses and contrasts them to highlight the future direction