文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > Geometric Visualization of Clusters Obtained from Fuzzy Clustering Algorithms
侵权投诉

Geometric Visualization of Clusters Obtained from Fuzzy Clustering Algorithms

Fuzzy-clustering methods, such as fuzzy k-means and Expectation Maximization, allow an object to be assigned to multiple clusters with different degrees of membership. However, the memberships that result from fuzzy-clustering algorithms are difficult to b

GeometricVisualizationofClustersObtainedfromFuzzy

ClusteringAlgorithms

LuisRueda andYuanquanZhang

Abstract

Fuzzy-clusteringmethods,suchasfuzzyk-meansandExpectationMaximization,allowanobjecttobeassignedtomultipleclusterswithdi erentdegreesofmembership.However,themembershipsthatresultfromfuzzy-clusteringalgorithmsaredi culttobeanalyzedandvisualized.Thememberships,usuallyconvertedto0-1values,arevisualizedusingparallelcoordinatesordi erentcolorshades.Inthispaper,weproposeanewapproachtovisualizefuzzy-clustereddata.Theschemeisbasedonageometricvisualization,andworksbygroupingtheobjectswithsimilarclustermembershipstowardstheverticesofahyper-tetrahedron.Theproposedmethodshowsclearadvantagesovertheexistingmethods,demonstratingitscapabilitiesforviewingandnavigatinginter-clusterrelationshipsinaspatialmanner.1Introduction

Manyclusteringalgorithmshavebeenproposedsofar,beingthemostimportantonesk-means

[21],fuzzyk-means[5],expectationmaximization(EM)[7],hierarchicalclustering[8],andself-organizingmaps[17].Someofthesemethodsgroupobjects,whichshareacertaindegreeofsimilarity,byassigningaprobabilityofmembershiptoeachcluster.Wecallthesemethods“fuzzy-clusteringalgorithms”.Inmolecularbiology,forexample,applyingfuzzy-clusteringtomicroarraydatabringstheadvantagethattheclusteringresultallowsagenetobeassignedtomorethanonecluster[5,9].Theproblem,however,ishowtoassigntheobjectstooneoftheclusters.Acommontechniquetodealwiththisistousea“cuto ”valueandassignanobjecttoacluster,ifitsmem-bershipprobabilityisabovethecuto .Ontheotherhand,visualizingthe“fuzzy”membershipofanobjectbelongingtodi erentclustersconstitutesinitselfanopenproblem.

Gaschetal.proposedamodi edhierarchicalvisualizationtoobservethefuzzy-clusteredgenesutilizingdi erentmembershipcuto values[10].Inthemembershiptable,eachobjectMemberoftheIEEE.DepartmentofComputerScience,UniversityofConcepcion,EdmundoLarenas215,Concepcion,VIIIRegion,Chile.Phone:+5641204305,Fax:+5641221770.E-mail:lrueda@inf.udec.cl. SchoolofComputerScience,UniversityofWindsor,401SunsetAve.,ON,N9B3P4,Canada.E-mail:kzhang@certicom.com.

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email