摘要： 介绍了关联规则的数据挖掘，对布尔关联规则中的经典算法——Apriori算法进行了分析和评价，指出了基本Apriori算法的不足，并提出具有广泛适应性的改进算法。此算法既提高在扫描数据库过程中的信息获取率，又及时剔除超集不是频繁项集的项集，进一步缩减项集的潜在规模，提高了频繁项集生成的效率。

关键词: 数据挖掘, 关联规则, Apriori算法, 候选项集, 频繁项集

Abstract: This paper focuses on the concepts and techniques of data mining association rules. Through an improvement of the Apriori algorithm, it can enhance the information-gain ratio in database scanning, identify more quickly the supersets that are not members of the large itemsets, and, thus, reduce the size of the large itemsets. It decreases the size of candidate itemsets by deleting some items in them in order to increase the efficiency of generating large itemsets.

中图分类号: 

• TP311.12