资源描述：
《Discovering Co-location Patterns from Spatial Datasets - A General Approach英文学习材料》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。
1、IEEETRANSACTIONSONKNOWLEDGEANDDATAENGINEERING1DiscoveringCo-locationPatternsfromSpatialDatasets:AGeneralApproachYanHuang,Member,IEEE,ShashiShekhar,Fellow,IEEE,andHuiXiong,StudentMember,IEEEAbstractGivenacollectionofbooleanspatialfeatures,theco-locationpatterndiscoverypr
2、ocessndsthesubsetsoffeaturesfrequentlylocatedtogether.Forexample,theanalysisofanecologydatasetmayrevealsymbioticspecies.Thespatialco-locationruleproblemisdifferentfromtheassociationruleproblem,sincethereisnonaturalnotionoftransactionsinspatialdatasetswhichareembededinc
3、ontinuousgeographicspace.Inthispaper,weprovideatransaction-freeapproachtomineco-locationpatternsbyusingtheconceptofproximityneighborhood.Anewinterestmeasure,aparticipationindex,isalsoproposedforspatialco-locationpatterns.Theparticipationindexisusedasthemeasureofprevalen
4、ceofaco-locationfortworeasons.First,thismeasureiscloselyrelatedtothecross- function,whichisoftenusedasastatisticalmeasureofinteractionamongpairsofspatialfeatures.Second,italsopossessesananti-monotonepropertywhichcanbeexploitedforcomputationalefciency.Furthermore,wedesi
5、gnanalgorithmtodiscoverco-locationpatterns.Thisalgorithmincludesanovelmulti-resolutionpruningtechnique.Finally,experimentalresultsareprovidedtoshowthestrengthofthealgorithmanddesigndecisionsrelatedtoperformancetuning.YanHuangiswiththeDepartmentofComputerScienceandEngine
6、ering,UniversityofNorthTexas,USA.E-mail:huangyan@unt.edu.ShashiShekhariswiththeDepartmentofComputerScienceandEngineering,UniversityofMinnesota,200UnionSteetSE,Minneapolis,MN55455,USA.E-mail:shekhar@cs.umn.edu.HuiXiongiswiththeDepartmentofComputerScienceandEngineering,Un
7、iversityofMinnesota,200UnionStreetSE,Minneapolis,MN55455,USA.E-mail:huix@cs.umn.edu.ManuscriptreceivedAugust26,2002;revisedJune17,2003;acceptedApril27,2004.CorrespondingAuthor:HuiXiong.Phone:612-626-8084.Fax:612-626-1596.IEEETRANSACTIONSONKNOWLEDGEANDDATAENGINEERING2Ind
8、exTermsCo-locationPatterns,SpatialAssociationRules,ParticipationIndex.I.INTRODUCTIONCo-locationpatternsreprese
