结合遗传算法的Apriori算法改进  

Improvement of Apriori algorithm based on genetic algorithm

作　　者：文武[1,2] 郭有庆 WEN Wu;GUO You-qing(Research Center of New Telecommunication Technology Applications,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Chongqing Information Technology Designing Limited Company,Chongqing 401121,China)

机构地区：[1]重庆邮电大学通信新技术应用研究中心,重庆400065 [2]重庆信科设计有限公司,重庆401121

出　　处：《计算机工程与设计》2019年第7期1922-1926,共5页Computer Engineering and Design

摘　　要：针对Apriori算法存在效率低、内存损耗大等问题,提出一种基于遗传算法来寻找频繁项集的(GNA)算法。结合Apriori算法和遗传算法的特点,设计 k 步挖掘过程,利用交叉算子产生候选项集和变异算子筛选频繁项集,避免多次扫描数据库的同时,减少冗余。实验结果表明,GNA算法相比Apriori算法,对稀疏数据集或稠密数据集,在挖掘频繁模式的数量及效率上都有显著提高。Aiming at the problems of low efficiency and large memory loss in Apriori algorithm,an algorithm (GNA) based on genetic algorithm was proposed to find frequent itemsets. Based on the characteristics of Apriori algorithm and genetic algorithm,the k -step mining process was designed. The candidate itemsets were generated using crossover operators and frequent itemsets were filtered using mutation operators,thus avoiding multiple scans of databases and reducing redundancy. Experimental results show that compared with the Apriori algorithm,the proposed algorithm can significantly improve the number and efficiency of mining frequent patterns for sparse or dense datasets.

关 键 词：关联规则 APRIORI算法 遗传算法 事务数据库 频繁模式 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]