一种为必威体育官网网址挖掘预处理数据的新方法.pdf

38 7 Vol.38 No.7 20117 Computer Science July 2011 ( 陕西师范大学计算机科学学院 西安710062) Apriori算法是数据挖掘中一个里程碑式的经典算法,在该算法的思想上衍生出了许多通过产生 繁项集来 导出关联规则的算法提出了一种基于数据项闭包的为必威体育官网网址数据挖掘进行数据预处理的全新方法该方法针对类 Apriori算法的特点和过程对不同特点的数据项进行不同的变换,使得挖掘请求方既能够正确地得到自己想要得到的 关于己方产品的信息,又无法正确地得到关于潜在的竞争对手的信息因此数据提供方在采用该方法预处理数据后, 能够提供有利于双方的数据,以达到双赢的目的 Apriori算法, 隐私保护, 数据项闭包,数据挖掘 TP391 A New Data Preprocessing Method for Privacypreserving Data Mining LIU Liang XIE Shuting LI Shundong ( School of Computer Science, Shaanx i Normal University, Xi an 710062, China) Abstract The Apriori algorithm is a milestone in the development of data mining. A number of other algorithms, which generate association rules by producing frequent itemsets, are derived from it.T his paper proposed a data preprocessing algorithm based on item closure, which is an absolutely new method for privacypreserving data mining. According to the characteristics and the processes of Apriori like algorithms,this method transforms different items in various ways. So the data mining applicant can only obtain the information of its own products correctly,but cannot obtain useful in formation regarding its potential competitors products.Hence, through the application of this method, the cooperated data mining between the data provider and data miner will benefit both sides,and obtains a winwin result. Keywords Apriori algorithm,Privacypreserving, Item closure, Data mining [ 5] , ; 1 , [ 6] R Agrawal 1993 [ 1] ; ( Data M