Abstract: Finding frequent itemset is a pivotal technology and stage in association rules mining application. Most studies adopt Apriori-like candidate set generation-and-test approach, but candidate set generation is still costly. This paper proposes an improved AprioriTid algorithm to improve the algorithmic executive efficiency, which is based on candidate set Lk address-mapping approach of Hash technology.

Key words: association rules, frequent item set, AprioriTid, Hash

摘要： 发现频繁项集是关联规则挖掘应用的关键，针对采用Apriori类的候选项目集生成-检验方法导致候选项目集产生的代价很高问题，该文提出一种基于散列的快速AprioriTid改进算法，在AprioriTid算法的基础上采用基于候选项Lk地址的哈希映射方法，提高了算法的执行效率。

关键词: 关联规则, 频繁项目集, AprioriTid算法, 散列

CLC Number: 

• TP301.6