摘要
对Apriori算法进行优化,提出了一种Z_Apriori算法。该算法在首次产生频繁项集时,扫描数据库并通过二进制编码串记录每个项目在事务里是否出现过,在每次进行计算迭代过程中无需再对数据库进行扫描,避免了对数据库的重复扫描,在系统性能和效率上较经典的Apriori算法有一定的改善。
Z_Apriori algorithm, which based on optimization of Apriori algorithm, is put forward in this paper. When generating frequent item sets for the first time, the algorithm can scan the database and record which project appears in the transaction by the binary code string and it no longer need to scan the database in each calculation and iteration process, which avoids scanning the database repeatedly. Compared with Apriori algorithm, it has a certain improvement about the performance and efficiency of the system.
出处
《三明学院学报》
2013年第2期27-31,共5页
Journal of Sanming University
关键词
关联规则
个性化推荐服务
频繁项集
association rules
personalized recommendation service
frequent item set
作者简介
曾姣艳，女，湖南郴州人，助教。研究方向：计算机应用。
