计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (4): 146-148.DOI: 10.3778/j.issn.1002-8331.2009.04.041

• 数据库、信号与信息处理 • 上一篇    下一篇

高效的关联规则快速更新算法

谭 军1,2,卜英勇2,杨 勃2   

  1. 1.中南林业科技大学 计算机学院,长沙 412006
    2.中南大学 机电工程学院,长沙 410083
  • 收稿日期:2008-07-28 修回日期:2008-08-30 出版日期:2009-02-01 发布日期:2009-02-01
  • 通讯作者: 谭 军

High-efficient fast updating algorithm for mining association rules

TAN Jun1,2,BU Ying-yong2,YANG Bo2   

  1. 1.College of Computer Science,Central South University of Forestry and Technology,Changsha 412006,China
    2.College of Mechanical & Electrical Engineering,Central South University,Changsha 410083,China
  • Received:2008-07-28 Revised:2008-08-30 Online:2009-02-01 Published:2009-02-01
  • Contact: TAN Jun

摘要: 挖掘关联规则的两大经典算法Apriori和FP-tree算法都是以批处理方式处理所有事务。但在实际应用中,新事务频繁地出现,这就需要不断更新关联规则。为了提高更新效率,有效减少扫描原数据库的次数,基于次频繁项的概念,在快速更新频繁模式树(FUFP-tree)算法的基础上,提出了一种改进的算法。实验结果表明新算法具有良好的性能。

关键词: 关联规则, Apriori算法, FP-tree算法, 快速更新频繁模式树, 次频繁项

Abstract: Apriori and FP-tree algorithm needed to process all transactions in a batch way.But in real applications,new transactions are usually inserted into databases.For improving efficiency of updating and decreasing the times of scanning original database,a new algorithm is proposed based on the concept of pre-large itemsets,which is at the base of FUFP-tree algorithm.Experimental results show that it has a good performance.

Key words: association rules, Apriori algorithm, FP-tree algorithm, FUFP-tree, pre-large itemsets