一种挖掘加权频繁项集的改进算法  

Improved algorithm for mining weighted frequent itemsets

作　　者：李彦伟[1] 戴月明[1] 王金鑫[1] 

机构地区：[1]江南大学信息工程学院,江苏无锡214122

出　　处：《计算机工程与应用》2011年第15期165-167,共3页Computer Engineering and Applications

摘　　要：分析了New-Apriori和MWFI(Mining Weighted Frequent Itemsets)算法之不足,提出了一种挖掘加权频繁项集的New-MWFI算法。该算法按属性的权值对事务进行分类,并依次求出每个类别内的加权频繁项集。由于每个类别内的频繁项集满足Apriori性质,因而可以利用Apriori算法或其他改进算法进行挖掘,从而克服了原来算法的不合理和效率低下的缺陷。实验表明该算法能更有效地从数据集中挖掘出加权频繁项集。The shortages of the New-Apriori and Mining Weighted Frequent Itemsets （MWFI） are analyzed, and the New-MWFI algorithm for mining weighted frequent itemsets is proposed.In this algorithm the transactions are classified according to the item＇s weight and the weighted frequent itemsets are mined within each category in turn.Since the frequent itemsets of each category satisfy the Apriori＇s property, the Apriori algorithm or other improved algorithms can be used, thus the deficiencies of the original algorithms can be overcome successfully.Experiments show that the new algorithm is more effective in mining the weighted frequent itemsets from the dataset.

关 键 词：数据挖掘 加权关联规则 加权频繁项集 New-MWFI算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]