Fuzzy Modeling and Genetic Algorithms for Data Mining and Exploration TheMorganKaufmannSeriesinDataManagementSystems SeriesEditor:JimGray,MicrosoftResearch FuzzyModelingandGeneticAlgorithmsforDataMiningandExploration EarlCox DataModelingEssentials,ThirdEdition GraemeC.SimsionandGrahamC.Witt Location-BasedServices JochenSchillerandAgnèsVoisard DatabaseModelingwithMicrosft(cid:1)VisioforEnterpriseArchitects TerryHalpin,KenEvans,PatrickHallock,BillMaclean DesigningData-IntensiveWebApplications StephanoCeri,PieroFraternali,AldoBongio,MarcoBrambilla,SaraComai,andMaristellaMatera MiningtheWeb:DiscoveringKnowledgefromHypertextData SoumenChakrabarti AdvancedSQL:1999—UnderstandingObject-RelationalandOtherAdvancedFeatures JimMelton DatabaseTuning:Principles,Experiments,andTroubleshootingTechniques DennisShashaandPhilippeBonnet SQL:1999—UnderstandingRelationalLanguageComponents JimMeltonandAlanR.Simon InformationVisualizationinDataMiningandKnowledgeDiscovery EditedbyUsamaFayyad,GeorgesG.Grinstein,andAndreasWierse TransactionalInformationSystems:Theory,Algorithms,andPracticeofConcurrencyControl andRecovery GerhardWeikumandGottfriedVossen SpatialDatabases:WithApplicationtoGIS PhilippeRigaux,MichelScholl,andAgnesVoisard InformationModelingandRelationalDatabases:FromConceptualAnalysistoLogicalDesign TerryHalpin ComponentDatabaseSystems EditedbyKlausR.DittrichandAndreasGeppert ManagingReferenceDatainEnterpriseDatabases:BindingCorporateDatatotheWiderWorld MalcolmChisholm DataMining:ConceptsandTechniques JiaweiHanandMichelineKamber UnderstandingSQLandJavaTogether:AGuidetoSQLJ,JDBC,andRelatedTechnologies JimMeltonandAndrewEisenberg Database:Principles,Programming,andPerformance,SecondEdition PatrickandElizabethO’Neil TheObjectDataStandard:ODMG3.0 EditedbyR.G.G.CattellandDouglasK.Barry DataontheWeb:FromRelationstoSemistructuredDataandXML SergeAbiteboul,PeterBuneman,andDanSuciu DataMining:PracticalMachineLearningToolsandTechniqueswithJavaImplementations IanWittenandEibeFrank JoeCelko’sSQLforSmarties:AdvancedSQLProgramming,SecondEdition JoeCelko JoeCelko’sDataandDatabases:ConceptsinPractice JoeCelko DevelopingTime-OrientedDatabaseApplicationsinSQL RichardT.Snodgrass WebFarmingfortheDataWarehouse RichardD.Hackathorn DatabaseModeling&Design,ThirdEdition TobyJ.Teorey ManagementofHeterogeneousandAutonomousDatabaseSystems EditedbyAhmedElmagarmid,MarekRusinkiewicz,andAmitSheth Object-RelationalDBMSs:TrackingtheNextGreatWave,SecondEdition MichaelStonebrakerandPaulBrown,withDorothyMoore ACompleteGuidetoDB2UniversalDatabase DonChamberlin UniversalDatabaseManagement:AGuidetoObject/RelationalTechnology CynthiaMaroSaracco ReadingsinDatabaseSystems,ThirdEdition EditedbyMichaelStonebrakerandJosephM.Hellerstein UnderstandingSQL’sStoredProcedures:ACompleteGuidetoSQL/PSM JimMelton PrinciplesofMultimediaDatabaseSystems V.S.Subrahmanian PrinciplesofDatabaseQueryProcessingforAdvancedApplications ClementT.YuandWeiyiMeng AdvancedDatabaseSystems CarloZaniolo,StefanoCeri,ChristosFaloutsos,RichardT.Snodgrass, V.S.Subrahmanian,andRobertoZicari PrinciplesofTransactionProcessing PhilipA.BernsteinandEricNewcomer UsingtheNewDB2:IBMsObject-RelationalDatabaseSystem DonChamberlin DistributedAlgorithms NancyA.Lynch ActiveDatabaseSystems:TriggersandRulesForAdvancedDatabaseProcessing EditedbyJenniferWidomandStefanoCeri MigratingLegacySystems:Gateways,Interfaces,&theIncrementalApproach MichaelL.BrodieandMichaelStonebraker AtomicTransactions NancyLynch,MichaelMerritt,WilliamWeihl,andAlanFekete QueryProcessingforAdvancedDatabaseSystems EditedbyJohannChristophFreytag,DavidMaier,andGottfriedVossen TransactionProcessing:ConceptsandTechniques JimGrayandAndreasReuter BuildinganObject-OrientedDatabaseSystem:TheStoryofO2 EditedbyFrançoisBancilhon,ClaudeDelobel,andParisKanellakis DatabaseTransactionModelsforAdvancedApplications EditedbyAhmedK.Elmagarmid AGuidetoDevelopingClient/ServerSQLApplications SetragKhoshafian,ArvolaChan,AnnaWong,andHarryK.T.Wong TheBenchmarkHandbookforDatabaseandTransactionProcessingSystems,SecondEdition EditedbyJimGray CamelotandAvalon:ADistributedTransactionFacility EditedbyJeffreyL.Eppinger,LilyB.Mummert,andAlfredZ.Spector ReadingsinObject-OrientedDatabaseSystems EditedbyStanleyB.ZdonikandDavidMaier Fuzzy Modeling and Genetic Algorithms for Data Mining and Exploration Earl Cox AMSTERDAM•BOSTON•HEIDELBERG•LONDON NEW YORK•OXFORD•PARIS•SANDIEGO SANFRANCISCO•SINGAPORE•SYDNEY•TOKYO MorganKaufmannPublishersisanimprintofElsevier PublishingDirector DianeD.Cerra AcquisitionsEditor RickAdams PublishingServicesManager SimonCrump ProjectManager DanStone DevelopmentalEditor EmiliaThiuri EditorialCoordinators CorinaDermanandMonaBuehler CoverDesign YvoRiezebosDesign CoverImage GettyImages Composition CEPHAImagingPvt.Ltd. TechnicalIllustration DartmouthPublishingInc. Copyeditor DarilBentley Proofreader PhyllisCoyneProofreadingServices Indexer LucieHaskins InteriorPrinter TheMaple-VailBookManufacturingGroup CoverPrinter PhoenixColor MorganKaufmannPublishersisanimprintofElsevier. 500SansomeStreet,Suite400,SanFrancisco,CA94111 Thisbookisprintedonacid-freepaper. ©2005byElsevierInc.Allrightsreserved. Designationsusedbycompaniestodistinguishtheirproductsareoftenclaimedastrademarks orregisteredtrademarks.InallinstancesinwhichMorganKaufmannPublishersisawareof aclaim,theproductnamesappearininitialcapitalorallcapitalletters.Readers,however, shouldcontacttheappropriatecompaniesformorecompleteinformationregarding trademarksandregistration. Nopartofthispublicationmaybereproduced,storedinaretrievalsystem,ortransmittedin anyformorbyanymeans—electronic,mechanical,photocopying,scanning,orotherwise— withoutpriorwrittenpermissionofthepublisher. PermissionsmaybesoughtdirectlyfromElsevier’sScience&TechnologyRightsDepartment inOxford,UK:phone:(+44)1865843830,fax:(+44)1865853333,e-mail:permissions@ elsevier.com.uk.Youmayalsocompleteyourrequeston-lineviatheElsevierhomepage (http://elsevier.com)byselecting“CustomerSupport”andthen“ObtainingPermissions.” LibraryofCongressCataloging-in-PublicationData APPLICATIONSUBMITTED ISBN:0-12-194275-9 ForinformationonallMorganKaufmannpublications, visitourWebsiteatwww.mkp.comorwww.books.elsevier.com PrintedintheUnitedStatesofAmerica 05 06 07 08 09 5 4 3 2 1 Contents Preface xiii ObjectivesandAudience xiv OrganizationoftheBook xv AlgorithmDefinitionsandExamples xv Acknowledgments xvii Introduction xix TheModernConnectedWorld xix TheAdventofIntelligentModels xix FuzzyLogicandGeneticAlgorithms xxi Part I Concepts and Issues ■■■ Chapter 1 Foundations and Ideas 3 1.1 EnterpriseApplicationsandAnalysisModels 4 1.2 DistributedandCentralizedRepositories 8 1.3 TheAgeofDistributedKnowledge 11 1.4 InformationandKnowledgeDiscovery 12 1.5 DataMiningandBusinessModels 18 1.6 FuzzySystemsforBusinessProcessModels 23 1.7 EvolvingDistributedFuzzyModels 25 1.8 ASampleCase:EvolvingaModelforCustomerSegmentation 26 1.9 Review 29 vii viii ■ Contents ■■■ Chapter 2 Principal Model Types 31 2.1 ModelandEventStateCategorization 34 2.2 ModelTypeandOutcomeCategorization 35 2.3 Review 36 ■■■ Chapter 3 Approaches to Model Building 37 3.1 OrdinaryStatistics 37 3.2 NonparametricStatistics 38 3.3 LinearRegressioninStatisticalModels 40 3.4 NonlinearGrowthCurveFitting 44 3.5 ClusterAnalysis 47 3.6 DecisionTreesandClassifiers 48 3.7 NeuralNetworks 51 3.8 FuzzySQLSystems 52 3.9 RuleInductionandDynamicFuzzyModels 55 3.10 Review 62 FurtherReading 63 Part II Fuzzy Systems ■■■ Chapter 4 Fundamental Concepts of Fuzzy Logic 67 4.1 TheVocabularyofFuzzyLogic 68 4.2 Boolean(Crisp)Sets:TheLawofBivalence 72 4.3 FuzzySets 76 4.4 Review 93 FurtherReading 94 ■■■ Chapter 5 Fundamental Concepts of Fuzzy Systems 95 5.1 TheVocabularyofFuzzySystems 96 Contents ■ ix 5.2 FuzzyRule-basedSystems:AnOverview 100 5.3 VariableDecompositionintoFuzzySets 104 5.4 AFuzzyKnowledgeBase:TheDetails 113 5.5 TheFuzzyInferenceEngine 119 5.6 InferenceEngineApproaches 122 5.7 RunningaFuzzyModel 124 5.8 Review 147 ■■■ Chapter 6 Fuzzy SQL and Intelligent Queries 149 6.1 TheVocabularyofRelationalDatabasesandQueries 150 6.2 BasicRelationalDatabaseConcepts 156 6.3 StructuredQueryLanguageFundamentals 159 6.4 PrecisionandAccuracy 161 6.5 WhyWeSearchDatabases 162 6.6 ExpandingtheQueryScope 165 6.7 FuzzyQueryFundamentals 169 6.8 MeasuringQueryCompatibility 180 6.9 ComplexQueryCompatibilityMetrics 184 6.10 CompatibilityThresholdManagement 187 6.11 FuzzySQLProcessFlow 188 6.12 FuzzySQLExample 193 6.13 EvaluatingFuzzySQLOutcomes 200 6.14 Review 204 FurtherReading 205 ■■■ Chapter 7 Fuzzy Clustering 207 7.1 TheVocabularyofFuzzyClustering 208 7.2 PrinciplesofClusterDetection 210 7.3 SomeGeneralClusteringConcepts 211 7.4 CrispClusteringTechniques 218 7.5 FuzzyClusteringConcepts 223 7.6 Fuzzyc-MeansClustering 228 7.7 FuzzyAdaptiveClustering 248