在单向FP-tree上挖掘频繁闭项集  

Mining frequent closed itemsets in unidirectional FP-tree

作　　者：王现君[1] 宋晶晶[2] 姜保庆[1] 

机构地区：[1]河南大学数据与知识工程研究所,河南开封475004 [2]清远职业技术学院信息科技学院,广东清远511510

出　　处：《计算机工程与应用》2008年第10期150-153,237,共5页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of China under Grant No.60474022);河南省高校杰出科研人才创新工程项目(No.2007KYCX018)

摘　　要：频繁闭项集提供了频繁项集的一种完整的、最小表示。针对稠密数据集,提出一种基于单向FP-tree的频繁闭项集挖掘算法Unid_FP-FCI。该算法在挖掘过程中只生成被约束子树,而它是一种虚拟的树结构,在原有的单向FP-tree基础上用三个很小的数组来表示,因而避免了以往算法需递归构造条件FP-tree来计算频繁闭项集的弊端,极大地降低了内存空间和时间开销,提高了挖掘效率。Frequent closed itemsets provide a minimal representation of frequent itemsets without losing their support information. This paper proposes an efficient algorithm Unid_FP-FCI for mining the complete set of frequent closed itemsets in a unidirectional FP-tree.Because in process of mining only generate constrained sub-trees consisting of three small an＇ays,which is pseudo tree structure based on the originally unidirectional FP-tree,the flaw is avoided in former algorithms which need to generate lots of conditional FP-trees for finding frequent closed itemsets recursively.Reducing the space and time consumption to a great extent,then the algorithm improve mining efficiency.

关 键 词：数据挖掘 频繁项集 频繁闭项集 单向FP-tree 被约束子树 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]