【摘要】 传统的关联规则Apriori算法在产生频繁项集的过程中,需要多次扫描事务数据库以及多次扫描频繁项集,从而造成算法性能下降.为了减少扫描事务数据库以及频繁项集的次数,在生成的候选k项集中,除了存储项集item-set以及支持度计数count之外,加入事务标识符列表Tidlist属性,在生成频繁k项集时,可以直接通过Tid-list的交集得出事务标识符列表以及项集的计数,不需要去扫描事务数据库,从而可以有效地提高算法的性能.文中提出了一种改进的关联规则挖掘模型以及关联规则挖掘算法I-Apriori算法.实验证明,I-Apriori算法相比Apriori算法的执行时间有明显改进.更多还原

【Abstract】 In the process of generating frequent itemsets,the traditional association rule Apriori algorithm needs to scan transaction databases and frequent itemsets many times,resulting in the decline of algorithm performance. In order to reduce the number of scanning the transaction database and frequent itemsets,when generating the candidate k-itemsets,transaction identifier list Tid-list is stored besides item-set and count. When generating frequent k-itemsets,we can get the list of transaction identifiers and the count of frequent itemsets directly through the intersection of Tid-list,and do not need to scan transaction database,which can effectively improve the performance of the algorithm. In this paper,an improved association rule mining model and association rule mining algorithm I-Apriori algorithm are proposed. The experiment shows that the I-Apriori algorithm has a significant improvement compared with the execution time of the Apriori algorithm.更多还原