资源描述：
《An expert system for detection of breast cancer based on association rules》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。
1、ExpertSystemswithApplications36(2009)3465–3469ContentslistsavailableatScienceDirectExpertSystemswithApplicationsjournalhomepage:www.elsevier.com/locate/eswaAnexpertsystemfordetectionofbreastcancerbasedonassociationrulesandneuralnetworka,*bMuratKarabatak,M.Cev
2、detInceaFıratUniversity,DepartmentofElectronicsandComputerScience,23119Elazig,TurkeybFıratUniversity,DepartmentofElectric-ElectronicsEngineering,23119Elazig,TurkeyarticleinfoabstractKeywords:Thispaperpresentsanautomaticdiagnosissystemfordetectingbreastcancerb
3、asedonassociationrulesAssociationrules(AR)andneuralnetwork(NN).Inthisstudy,ARisusedforreducingthedimensionofbreastcancerdata-NeuralnetworkbaseandNNisusedforintelligentclassification.TheproposedAR+NNsystemperformanceiscomparedAutomaticdetectionwithNNmodel.Thedi
4、mensionofinputfeaturespaceisreducedfromninetofourbyusingAR.IntestBreastcancerstage,3-foldcrossvalidationmethodwasappliedtotheWisconsinbreastcancerdatabasetoevaluatetheproposedsystemperformances.Thecorrectclassificationrateofproposedsystemis95.6%.Thisresearchde
5、monstratedthattheARcanbeusedforreducingthedimensionoffeaturespaceandproposedAR+NNmodelcanbeusedtoobtainfastautomaticdiagnosticsystemsforotherdiseases.Ó2008ElsevierLtd.Allrightsreserved.1.Introductionbreastcancer(Chouaetal.,2004).Inliterature,radiologistsshowc
6、onsiderablevariationininterpretingamammography(ElmoreDataclassificationprocessusingknowledgeobtainedfrometal.,1994).Fineneedleaspirationcytology(FNAC)isalsowidelyknownhistoricaldatahasbeenoneofthemostintensivelystudiedadoptedinthediagnosisofbreastcancer.But,th
7、eaveragecorrectsubjectsinstatistics,decisionscienceandcomputerscience.IthasidentificationrateofFNACisonly90%.So,itisnecessarytodevelopbeenappliedinproblemsofmedicine,socialsciencemanagementbetteridentificationmethodtorecognizethebreastcancer.Statis-andengineeri
8、ng.Variableproblemssuchasdiseasediagnosis,im-ticaltechniquesandartificialintelligencetechniqueshavebeenagerecognition,andcreditevaluationusingclassificationtech-usedtopredictthebreastcancer
