基于BTA算法的数据流频繁项集挖掘  

Frequent itemsets mining algorithm for data stream based on BTA algorithm

作　　者：文凯[1,2,3] 耿小海 许萌萌 WEN Kai;GENG Xiao-hai;XU Meng-meng(School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Research Center of New Telecommunication Technology,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Chongqing Information Technology Designing Limited Company,Chongqing 401121,China)

机构地区：[1]重庆邮电大学通信与信息工程学院,重庆400065 [2]重庆邮电大学通信新技术应用研究中心,重庆400065 [3]重庆信科设计有限公司,重庆401121

出　　处：《计算机工程与设计》2020年第8期2226-2230,共5页Computer Engineering and Design

摘　　要：针对传统数据流频繁项集挖掘算法在挖掘频繁k-项集时会有候选项集产生,在有新的数据流到来时的数据更新以及频繁项集支持度更新的效率不高,造成挖掘的时间和空间效率不高等一系列问题,提出一种高效的数据流频繁项集挖掘算法BTA(bit table with and algorithm)算法。将数据高效压缩进位表中,对窗口更新采用取余覆盖;在频繁k-项集的挖掘采用与操作避免候选项集产生;在支持度更新采用加减运算得到数据更新后的支持度。实验结果表明,该算法在时间和空间效率上均有良好效果。In view of the inefficiency of the traditional frequent itemset mining algorithms in mining frequent k-itemsets,such as the generation of candidate itemsets,the updating of data when new data streams arrive,and the updating of support degree of frequent itemsets,which results in the inefficiency of mining time and space,an efficient algorithm,namely BTA(bit table with and algorithm),was proposed for mining frequent itemset of data streams.Data were compressed efficiently in carry table,redundancy coverage was used for window updating,candidate itemsets generation were avoided using and manipulating in frequent k-itemsets mining,and addition and subtraction operation were used for support updating to obtain support after data updating.Experimental results show that the algorithm has good performances both in time and space efficiency.

关 键 词：数据流 频繁项集 压缩位表 与操作 数据挖掘 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]