摘要: 频繁模式挖掘的研究对象包括事务、序列、树和图。该文提出用模式增长方法在无序树构成的森林中挖掘嵌入频繁子树。利用规范化方法实现用唯一的形式表现无序树,根据待增长模式的拓扑结构确定其增长点并构造相应的投影库,将挖掘频繁子树模式问题转化为在各个投影库中寻找频繁节点的问题。
关键词:
频繁模式,
频繁子树,
无序树,
嵌入式子树
Abstract: Frequent patterns mining involves mining transactions, sequences, trees and graphs. This paper presents an efficient pattern growth algorithm for mining frequent embedded subtrees in a forest composed of unordered trees. It uses a canonical method to represent unordered trees in a unique way. It creates a projection database for every growing point of the pattern to grow. The problem is transformed from mining frequent trees to find frequent nodes in the projection database.
Key words:
frequent pattern,
frequent subtree,
unordered tree,
embedded subtree
中图分类号:
刘 波;杨 燕. 无序嵌入式频繁子树挖掘算法[J]. 计算机工程, 2009, 35(3): 51-53.
LIU Bo; YANG Yan. Mining Algorithm for Unordered Embedded Frequent Subtree[J]. Computer Engineering, 2009, 35(3): 51-53.