计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (30): 138-141.DOI: 10.3778/j.issn.1002-8331.2010.30.041

• 数据库、信号与信息处理 • 上一篇    下一篇

多属性变化的增量关联规则更新研究

曹增明,于书举,贾豪杰   

  1. 北京工业大学 计算机学院,北京 100124
  • 收稿日期:2009-03-26 修回日期:2009-05-21 出版日期:2010-10-21 发布日期:2010-10-21
  • 通讯作者: 曹增明

Research on incremental updating for mining association rules based on change in multiple attributes

CAO Zeng-ming,YU Shu-ju,JIA Hao-jie   

  1. School of Computer Science,Beijing University of Technology,Beijing 100124,China
  • Received:2009-03-26 Revised:2009-05-21 Online:2010-10-21 Published:2010-10-21
  • Contact: CAO Zeng-ming

摘要: 在支持度和事务库发生变化时,如何有效地更新关联规则的问题是目前数据挖掘研究的热点。但当事务库中的属性发生变化时,如何高效地更新关联规则的问题一直都没有引起研究人员的重视。ACA+和ACA-算法对单属性增减后的关联规则进行了研究,在此基础上,提出了解决多属性增减的增量关联规则更新算法MACA+和MACA-。通过建立事务-属性矩阵有效地解决了该问题。

关键词: 多属性, 数据挖掘, 关联规则, 基于多属性变化, 增量式关联规则更新

Abstract: When support and transaction database are changing,the problem of how to effectively update association rules is a hot topic in data mining research.However the problem of how to effectively update association rules after a change of the attributes in transaction database has not too much attention.ACA+ and ACA-algorithm do researches on signal attribute change of it.Based on this,the paper proposes incremental updating algorithms for mining association rules based on the change of multiple attributes,which are MACA+ and MACA-.By structuring transaction-attribute matrix,it can effectively solve the problem.

Key words: multiple attribute, data mining, association rules, multiple attribute change, incremental association rules updating

中图分类号: