摘要
针对MASK算法的不足,将随机响应技术与关联规则挖掘算法相结合,提出一个多参数随机扰动算法—MRD算法。当以不同的随机参数对数据集进行处理时,可以实现对原始数据的干扰或隐藏,解决了单一使用数据干扰策略和数据隐藏策略的缺陷,有效地提高了算法的隐私保护度。在此基础上,给出了在伪装后的数据集上生成频繁项集的挖掘算法。最后,通过具体实例验证,证明了当随机参数选择合适时,MRD算法的隐私性和准确性均优于原算法。
In view of the insufficiency of MASK algorithm,randomized response technology and association rule mining algorithm are integrated and a multi-parameters randomized disturb algorithm is proposed,which is called MRD algorithm.When the data sets are processed with different randmn parameters,the original data can be disturbed and hidden,and the defects of the simplex using of data diturb and data hiding strategy are solved, and the privacy-preserving degree of the algorithm is improved effectively.On this basis,the algorithm of generating frequent items from transformed data sets is proposed.Finally,through specific certification of examples,it can be proved that when the random parameters are choosen suitably,the privacy and accuracy of MRD algorithm are both better than the original algorithm.
出处
《计算机工程与应用》
CSCD
北大核心
2009年第26期126-130,共5页
Computer Engineering and Applications
基金
黑龙江省自然科学基金No.F0310
哈尔滨工程大学基础研究基金项目(No.HEUF04091)~~
关键词
数据挖掘
关联规则
频繁项集
隐私保护
随机响应
data mining
association rule
frequent itemset
privacy preservation
randomized response
