聚类趋势问题的研究综述  

Research for clustering tendency

作　　者：褚娜[1] 马利庄[1,2] 王彦[1] 

机构地区：[1]上海交通大学电子信息与电气工程学院计算机科学与工程系,上海200240 [2]上海中医药大学信息科学与技术中心,上海201203

出　　处：《计算机应用研究》2009年第3期801-803,822,共4页Application Research of Computers

基　　金：国家“973”计划资助项目(2006CB504801);上海市科委世博科技专项资助项目(06dz05815)

摘　　要：聚类算法的性能与数据集的结构是密切相关的,虽然目前已经研究出了很多聚类算法,但没有普遍适用的万能聚类算法,欠缺对数据集结构的有效解释。对聚类分析过程中重要的关键性问题,即聚类趋势问题进行了系统性的研究,从统计检验、可视化分析等角度给予了讨论,为数据集的无监督聚类分析提供了合理和有效的前期分析工具。It is closely related between the performance of clustering algorithms and the structure of data sets.No methods were good enough for all types of data,nor were all methods equally applicable to all problems,and were short of reasonable interpretation to data structure.And systematic researched the clustering tendency,which was one of key problem about clustering analysis.This paper discussed it based on statistic tests,visual analysis and so on,and proved that it could present reasonable and effective anal...

关 键 词：聚类趋势 聚类分析 统计检验 可视化评估 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]