集群环境下企业应用系统的关联规则算法研究  

On Association Rules Algorithm of Enterprise Application System in Cluster Environment

作　　者：赵向兵 张景安[1] ZHAO Xiang-bing;ZHANG Jing-an(School of Computer and Network Engineering,Shanxi Datong University,Datong Shanxi,037009)

机构地区：[1]山西大同大学计算机与网络工程学院,山西大同037009

出　　处：《山西大同大学学报（自然科学版）》2018年第6期31-33,81,共4页Journal of Shanxi Datong University(Natural Science Edition)

基　　金：山西省教育科学"十三五"规划[GH-18044];山西大同大学科研基金项目[2017K11][2015K2];山西大同大学教学改革创新项目[XJG2018231];大同市科学研究项目[JXW2017001]

摘　　要：大数据在各个领域的快速发展,推动着企业不断地发展新业务和创造新的发展模式,企业大数据的应用和挖掘,成为企业提高竞争力的关键因素之一。关联规则作为数据挖掘研究的主要领域,频繁模式的发现是提高关联规则挖掘效率的关键,随着数据量的不断增加,频繁模式发现过程存在I/O代价大和内存占有高等不足,本算法对数据集中事务项,采用MapReduce分布式编程模型,用两对Map和Reduce函数,实现了支持度计数和频繁项集生成,最终生成关联规则。最后,采用企业信息化调研数据,实验验证了该算法的可拓展性和可收缩性。With rapid development of big data in various fields,the enterprises has been pushed to develop new businesses and create new development models continuously.The application and excavation of big data has become one of the key factors for enterprises to improve their competitiveness.Association rules are the main research field of data excavation,and the discovery of frequent patterns is the key to improve the excavation efficiency of association rules.With the increase of data amount,the frequent pattern discovery process has the disadvantages of high I/O cost and high memory possession.In this algorithm,MapReduce distributed programming model and two pairs of Map and Reduce functions are used to realize the support count and frequent item set generation,and finally to generate association rules.Finally,using the enterprise information research data,the experiment verifies the scalability and shrinkability of the algorithm.

关 键 词：关联规则 频繁模式 MapReduce编程模型 企业大数据挖掘 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]