基于有向图的关联规则挖掘算法研究  

Research association rule mining algorithm on directed graph

作　　者：唐德权[1] 夏耀稳[1] 朱林立[1] 夏幼明[1] 

机构地区：[1]云南师范大学计算机科学与信息技术学院,云南昆明650092

出　　处：《云南大学学报（自然科学版）》2006年第S2期119-121,125,共4页Journal of Yunnan University(Natural Sciences Edition)

基　　金：云南省教育厅自然科学基金资助项目(5J0621D)

摘　　要：为了提高关联规则挖掘效率,提出了一种基于有向图的频繁项目集挖掘算法DGBFIG(Directed graph -based frequent itemsets generation)．该算法采用位矢量技术构造有向图,表示项与项之间的频繁关系,并在有向图的基础上递归产生频繁项集,从而只需扫描数据库2次,不产生候选集,从而大大提高了关联规则挖掘算法的效率．最后从空间和时间的复杂度分析了该算法的效率．To improving the efficiency of association ruler, it is proposed that a frequent itemsets mining algorithm based direct graph named DGBFIG(Directed graph-based frequent itemsets generation). Direct graph technique is set up by bit vector, which present the frequent relation between items. It generate frequent item set based on recursion of direct graph. So it only scan database twice to improving the efficiency of association ruler mining. The efficiency of time and space are analyzed by this method at last.

关 键 词：数据挖掘 有向图 频繁项集 位矢量 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]