J4

• • 上一篇    下一篇

基于FP-tree频集模式的FP-Growth算法对关联规则挖掘的影响

陆 楠1, 王 哲2, 周春光2   

  1. 1. 深圳大学信息工程学院, 深圳 518060; 2. 吉林大学计算机科学与技术学院, 长春 130012
  • 收稿日期:2002-08-26 修回日期:1900-01-01 出版日期:2003-04-26 发布日期:2003-04-26
  • 通讯作者: 周春光

The Effect of FP-Growth Algorithm Based on FP-tree Frequent Set Patterns on Asso ciation Rule Mining

LU Nan1, WANG Zhe2, ZHOU Chun-guang2   

  1. 1. College of Information Engineering, Shenzhen University, Shenzhen 518060, China; 2. College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2002-08-26 Revised:1900-01-01 Online:2003-04-26 Published:2003-04-26
  • Contact: ZHOU Chun-guang

摘要: 通过对两个有代表性的算法Apriori和FP-Growth的剖析, 说明频集模式挖掘的过程 , 比较有候选项集产生和无候选项集产生算法的特点, 并给出FP-tree结构的构造方法以 及对挖掘关联规则的影响, 提出了对算法的改进方法.

关键词: 数据挖掘, 关联规则, 频繁项集, 无候选项集

Abstract: An anatomy of two representative arithmetics of the Apriori and the FP -Growth explains the mining process of frequent-patterns item set. The improved method is put forward by comparing the arithmetic characteristics of candidate item set and non-candidate item set. The constructing method of FP-tree structure is provided and how it affects association rule mining is discussed.

Key words: data mining, association rule, frequent item set, non-candidate item set

中图分类号: 

  • TP391