作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (13): 56-57,6. doi: 10.3969/j.issn.1000-3428.2008.13.021

• 软件技术与数据库 • 上一篇    下一篇

基于回收技术的关联规则研究

刘 莹,景 波,黄 兵   

  1. (南京审计学院信息科学学院,南京 210029)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-07-05 发布日期:2008-07-05

Research on Association Rule Based on Recycle Technique

LIU Ying, JING Bo, HUANG Bing   

  1. (School of Information Science, Nanjing Audit University, Nanjing 210029)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-07-05 Published:2008-07-05

摘要: 关联规则的研究目前已经能够从含有缺失值的数据间建立关联性,但缺失值填充的完整性仍显不足。该文利用规则回收技术,以回收组合的方法将已往在挖掘过程中被删除掉的关联规则加以回收利用,从而可以获得更多的关联规则。这种以回收获得的组合式关联规则不仅能够提升缺失值的填充率和正确率,而且可以改进关联规则挖掘方法,降低挖掘时间及空间的复杂度。

关键词: 关联规则, 缺失值填充, 频繁k-项集

Abstract: The research on association rules can establish the relationship of data containing missing value. While the integrity on filling the missing value is lacking. This paper proposes a rule recycle technique, which reuses the rules deleted in previous data mining by reclaiming and composing them in order to obtain more association rules. The recycle combined association rules can promote the filling rate and accuracy of missing values. And improves association rules mining algorithm which reduces the complexity of time and space.

Key words: association rule, missing data imputation, frequent k-itemsets

中图分类号: