分布式环境基于相似度的关联规则挖掘模型的研究  

Similarity-based distributed data mining model research

作　　者：谈冉[1] 陆正球[1] 严新平[1] 

机构地区：[1]武汉理工大学计算机学院,武汉430063

出　　处：《计算机应用研究》2008年第3期695-697,701,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(60572015)

摘　　要：介绍了关联规则挖掘的基本原理和方法,详细分析了分布式关联规则挖掘算法并给出其模型;提出一种充分考虑数据源异构性、基于相似度的的分布式数据挖掘方法。实验证明该模型提高了挖掘的准确率。This paper introduced the basic theory and methods of association rules mining, analyzed and discussed the existing algorithms for mining distributed association rules and then put forward a new method for distributed data mining based on similarity which took well the heterogenous data source into account. Finally it realized the model. The experiment also proves the increased veracity.

关 键 词：数据挖掘 关联规则 相似度 分布式系统 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]