低支持度关联规则挖掘的一种算法  

An Algorithm for Mining Association Rules with Low Support

作　　者：卢世海[1] 齐雁[2] 

机构地区：[1]中原工学院计算机科学系,河南郑州450007 [2]武汉大学计算机学院,湖北武汉430072

出　　处：《中原工学院学报》2003年第2期57-59,共3页Journal of Zhongyuan University of Technology

摘　　要：　针对已有的对低支持度关联规则进行挖掘的算法中没有提出对具有多个相关项的关联规则进行挖掘的有效方法,本文提出一种能够对低支持度关联规则的多个相关项进行有效挖掘的方法.算法基于相似度来衡量各个相关项的关联程度,在已有算法的基础上增加了一次特殊的矩阵转换,从而将对项的相似度衡量方法进行了转换,转换后的矩阵可以基于Apriori性质来拓展多个相似相关项.算法在低支持度情况下具有较高的挖掘效率和良好的挖掘效果.算法还可以用来挖掘多个项之间的排斥规则.To overcome the limitation of existing algorithm on mining association rules composed of multiitems with low support , this paper proposes a new method which can mine association rules composed of multiitems with low support effectively. The algorithm given in this paper measures the extent of similarity of the items on the base of Similarity and an extra special matrix transform is introduced in this paper. The way of measuring the similarity of items is changed and the transformed matrix has the property Apriori, so we can then get more similar items than two. This algorithm can mine the association rules with low support effectively. It can also be used to obtain the mutual exclusion rules of multiitems.

关 键 词：数据挖掘 关联规则 APRIORI性质 算法 矩阵转换 相似度 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置] TP311.13[自动化与计算机技术—控制科学与工程]