【摘要】 关联规则挖掘是数据挖掘领域中最重要的研究问题之一。Apriori是关联规则挖掘的一种经典算法,它使用候选项集产生测试机制来找出所有满足用户最小支持度的项集,但它需要多次扫描数据库,会产生大量的候选项集。针对Apriori算法的不足,提出了一种基于混合型新的优化算法:Apriori-Mend算法。该算法从优化产生2项集、事务压缩等几个方面对Apriori算法进行优化,将散列技术应用于产生1项集和2项集,采用库优化策略和混合型存储结构,以节省空间和运算时间。实验结果表明,Apriori-Mend算法运行速度比Apriori算法有明显的提高。更多还原

【Abstract】 Finding association rules is one of the most popular problems in the field of data mining.Apriori is a well known algorithm for association rule mining.This process uses candidate generation and test method to find the itemsets that satisfy the minimum support threshold,but the process repeatedly scans the database and produces the plenty of candidates.Aiming at this insufficient of apriori algorithm,it proposes a new optimization algorithm called apriori-mend based on mixed structure optimization.Apriori-mend algorithm generates 2-items generation from optimizations,realizes transactions compression and so on.Apriori-mend uses hash structure to generate 1-items and 2-items,applies mixed structure optimization strategy to save time and space.The results show that apriori-mend algorithm is much faster than apriori algorithm.更多还原