用于短频繁项的隐私保护关联规则挖掘方法  

A Privacy Preserving Association Rules Mining Method for Short Frequent Itemsets

作　　者：张翠翠[1] 阮树骅[1] 

机构地区：[1]四川大学计算机学院,四川成都610065

出　　处：《电子科技》2016年第5期88-92,共5页Electronic Science and Technology

摘　　要：随着数据量的增长,隐私保护的问题也愈发突出,文中是介绍了目前数据挖掘过程中隐私保护相关的基本技术,提出了一种数据集中式分布下布尔数据集的关联规则的挖掘算法,此方法在实现了隐私保护的同时,通过与或运算实现了数据集的压缩。相关实验数据表明,该算法有效减少了挖掘时间,并保证了误差在可接受的范围之内。The explosion of data poses increasingly challenges on privacy preservation. This paper introduces basic technologies related to the privacy preservation in data mining,and puts forward a mining algorithm under association rule to deal with the Boolean data set distributed in a centralized manner. The method compresses the data set while preserving privacy,thus enormously reducing the time of data mining. Test results show that the mining time is significantly reduced with acceptable errors by this algorithm.

关 键 词：隐私保护 关联规则 压缩 数据挖掘 短频繁项集 

分 类 号：TP311.563[自动化与计算机技术—计算机软件与理论]