基于MapReduce的约束频繁项集挖掘算法  

Constrained frequent itemsets mining algorithm based on MapReduce

作　　者：施亮[1] 钱雪忠[1] 

机构地区：[1]江南大学物联网工程学院,江苏无锡214122

出　　处：《计算机工程与设计》2015年第10期2725-2728,2748,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(61103129;61202312);江苏省科技支撑计划基金项目(BE2009009)

摘　　要：传统的约束频繁项集挖掘方法无法实现对较大数据量的快速处理,针对该问题,结合分布式框架Hadoop的分布式计算优势,提出一种基于MapReduce的约束频繁项集挖掘算法。将一个完整的挖掘任务分成若干个相对独立的子任务,根据用户自定义的约束条件对子任务进行并行挖掘,提高算法的执行效率。实验结果表明,该算法具有较好的实用性和良好的扩展性。Traditional constrained frequent item set mining methods fail to achieve fast processing of large amounts of data.To solve these problems,a constraint frequent item sets mining algorithm based on MapReduce combining with distributed computing framework advantage of the distributed Hadoop.A complete mining task was split into several relatively independent subtasks,and subtasks were parallel mined based on user-defined constraints,so that the efficiency of the algorithm was improved.Experimental results show that the algorithm has good practicability and good scalability.

关 键 词：数据挖掘 MapReduce编程模型 约束频繁项集 频繁模式树 关联规则 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]