Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (11): 149-151.

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Improved Apriori algorithm

LIU Weixiao,CHEN Junli,QU Shifu,WAN Wanggen   

  1. Department of Communication and Information Engineering,Shanghai University,Shanghai 200072,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-04-11 Published:2011-04-11

一种改进的Apriori算法

刘维晓,陈俊丽,屈世富,万旺根   

  1. 上海大学 通信与信息工程学院,上海 200072

Abstract: The Apriori algorithm has some abuses,such as too many scans of the database,large load of system’s I/O and vast unrelated middle itemsets.This paper proposes an improved Apriori algorithm to overcome the abuses.The improved algorithm reduces the set of candidates and accelerates the speed of the algorithm by adding the interest items.Breaking the traditional steps of the algorithm to reduce the database scans and bring down the load of system’s I/O.The algorithm improves the readability of the strong association by constructing the model of the interest measure.Experimental results show that the algorithm can improve the speed and efficiency of operation effectively.

Key words: data mining, association rule, interest items, interest measure, model

摘要: 针对Apriori算法对数据库的扫描次数过多、系统的I/O负载大和产生大量的无关中间项集等弊端,提出了一种改进的Apriori算法。该算法通过加入用户兴趣项,减少候选项集的产生;打破了传统的算法实现步骤减少了数据库的扫描次数,降低了系统I/O负载;构建了用户兴趣度模型增加了算法生成强关联规则的可读性,提高了算法的效率。实验表明,改进的Apriori算法能有效地提高运行速度和效率。

关键词: 数据挖掘, 关联规则, 兴趣项, 兴趣度, 模型