基于案例的机遇发现决策支持技术及其系统的分析-case - based decision support technology of opportunity discovery and its systematic analysis.docx

hypergraph.Thenthegranularitydescriptionandcalculationmethodofhypergraphsystem is studied.Thirdly,towardscaseretrievalissue,existingsimilaritymeasuringmethedsare summarized,andamodelofsimilaritymetricsforhypergraphsystemisproposed,the computationalprocedureofwhichisconsistofcomparabilityanalysis,correspondence solving,andsimilaritydegreecomputing.Forthecorrespondencesolvingofverticesin hypergraphsystem,amaximalmatchingenumerationalgorithmforbipartitegraphbased oncomparabilityadjacencymatrixisproposed.Theusefulnessofourmodeland algorithmis verified by two groups of experiments.Forthly,directingatlearningofopportunitydiscoverycases,theissueofopportunity discoveryactivitypatternminingispresented.Byweightingactivitiesconcerning differentimportances,aopportunitydiscoveryactivitypatternminingalgorithmbasedon weightedsupport,wGSP algorithm,isproposed.Ontheresultsofactivitypatternmining, a similarity matching method for opportunity discovery activity sequences is studied.Fifthly,accordingtoproblemsolvingtheoriesandmethods,theproblemsolving basedopportunitydiscoverydecisionsupportprocessisstudied,includingtherecursionof informationassimilation,caseretrievalandcaserevisionstages.Theways,operatorsand methodsofinformationassimilation,factormodelandmodeofcaseretrieval,ways, operatorsandmethodsofcaserevisionarestudiedrespectively.Aheuristicalgorithmfor caserevisionbasedonk-starhypergraphextractionisproposed.Theworkingprocessof case-based opportunity discovery decision support is elaborated.Sixthly,theworkingrequirementsofdecisionsupportandworkingmechanismof DSSforopportunitydiscoveryisanalyzed.ByintroducingMulti-Agenttechnique,the structureandworkingprocessofconstitutedAgentsareinvestigated,andthecooperation modelofMulti-Agentforopportunitydiscoverydecisionsupportisproposed.The architectureandfunctionsofcase-basedopportunitydiscoverydecisionsupportsystem are presented, and the platformand an example of systemimplementation is shown.Finally,thecontentsandconclusionsofthisdissertat