预览加载中，请您耐心等待几秒...

基于PAM聚类方法的RBF神经网络设计.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于PAM聚类方法的RBF神经网络设计 基于PAM聚类方法的RBF神经网络设计 Abstract RecurrentNeuralNetworks(RBF)havebeenwidelyusedinthefieldofpatternrecognitionandfeaturerecognitionbecauseoftheirexcellentperformanceinnonlinearmapping.Atthesametime,PAMclusteringalgorithmhasbeenwidelyusedindataclustering,whichcaneffectivelyachievethegoalofminimizingintra-clusterdistanceandinter-clusterdistance.Inthispaper,thePAMclusteringalgorithmiscombinedwiththeRBFneuralnetworktodesignanefficientpatternrecognitionsystem. Introduction Withthecontinuousdevelopmentofscienceandtechnology,datahasbecomeanincreasinglyimportantresource.Inordertomakebetteruseofdataresources,itisnecessarytocarryoutdataminingandpatternrecognition.RBFneuralnetworkisakindoffeedforwardneuralnetworkbasedontheprinciplesofstatistics,mathematicsandcomputationallogicthatiswidelyusedinpatternrecognition.PAMclusteringalgorithmisaclassicclusteringalgorithmthatcaneffectivelyidentifytheclusteringofdata.Bycombiningthetwo,wecandesignanefficientpatternrecognitionsystem. TheprincipleofPAMclusteringalgorithm PAMclusteringalgorithm,alsoknownaspartitioningclusteringalgorithm,isawidelyusedclusteringalgorithmindatamining.Itscoreidealiesinthecomprehensiveconsiderationofintra-clusterdistancesandinter-clusterdistances,andidentifyingtheclusteringofdatabyiterativelyupdatingandoptimizingtheclusteringstructure. TheprincipleofRBFneuralnetwork RBFneuralnetworkisakindoffeedforwardneuralnetwork.Itsbasicstructureismainlycomposedofinputlayer,hiddenlayerandoutputlayer.Theinputlayeracceptstheinputdata,andthehiddenlayeriscomposedofaseriesofradialbasisfunctions.Theoutputlayerisresponsibleforoutputtingtheresults.RBFneuralnetworkhastheadvantagesoffastconvergence,goodgeneralizationabilityandstrongnonlinearmappingability. ThedesignofRBFneuralnetworkbasedonPAMclusteringalgorithm InthedesignofRBFneuralnetworkbasedonPAMclusteringalgorithm,thePAMclusteringalgorithmisappliedtothetrainingset,andtheclusteringstructureofthetrainingsetisidentified.Then,accordingtotheclusteringstructure,thecenter