简介：摘要双聚类（Biclustering）算法在数据挖掘中是一个新兴的算法，对于矩阵类型的数据，其聚类效果很好。本文浅述了双聚类算法的基本特点，并提出了用迭代的双聚类算法对未知的数据进行分类，并对一组数据进行了测试，其分类表现不错。

简介：Thispaperpresentsanewfuzzymultiplecriteria(bothqualitativeandquantitative)decision-making(MCDM)methodbasedonfuzzyrelationaldegreeanalysis.Theconceptsoffuzzysettheoryareusedtoconstructaweightedsuitabilitydecisionmatrixtoevaluatetheweightedsuitabilityofdifferentalternativesversusvariouscriteria.Thepositiveidealsolutionandnegativeidealsolutionarethenobtainedbyusingamethodofrankingfuzzynumbers,andthefuzzyrelationaldegreesofdifferentalternativesversuspositiveidealsolutionandnegativeidealsolutionarecalculatedbyusingtheproposedarithmetic.Finally,therelativerelationaldegreesofvariousalternativesversuspositiveidealsolutionarerankedtodeterminethebestalternative.Anumericalexampleisprovidedtoillustratetheproposedmethodattheendofthispaper.

简介：ＦｕｚｚｙＩｎｔｅｒｖａｌＶａｌｕｅＬｏｇｉｃａｎｄＦｕｚｚｙＤｉｓｔｒｂｕｔｅｄＶａｌｕｅＬｏｇｉｃTXＦｕｚｚｙＩｎｔｅｒｖａｌＶａｌｕｅＬｏｇｉｃａｎｄＦｕｚｚｙＤｉｓｔｒｉｂｕｔｅｄＶａｌｕｅＬｏｇｉｃＨｅＸｉｎｇｕｉＢｅｉｊｉｎｇＩｎｓｔｉｔ...

简介：Syllogisticfuzzyreasoningisintroducedintofuzzysystem,andthenewCascadedFuzzySystem(CFS)ispresented.ThethoroughlytheoreticalanalysisandexperimentalresultsshowthatsyllogisticfuzzyreasoningismorerobustthanallotherimplicationinferencesfornoisedataandthatCFShasbetterrobustnessthanconventionalfuzzysystems,whichprovidethesolidfoundationforCFS'spotentialapplicationinfuzzycontrolandmodelingandsoon.

简介：

简介：Thispaperproposesanewneuralfuzzyinferencesystemthatmainlyconsistsoffourparts.Thefirstpartisabouthowtouseneuralnetworktoexpresstherelationwithinafuzzyrule.Thesecondpartisthesimplificationofthefirstpart,andexperimentsshowthatthesesimplificationswork.Onthecontrarytothesecondpart,thethirdpartistheenhancementofthefirstpartanditcanbeusedwhenthefirstpartcannotworkverywellinthefuzzyinferencealgorithm,whichwouldbeintroducedinthefourthpart.Finally,thefourthpart"neuralfuzzyinferencealgorithm"isbeenintroduced.Itcaninferencethenewmembershipfunctionoftheoutputbasedonpreviousfuzzyrules.Theaccuracyofthefuzzyinferencealgorithmisdependentonneuralnetworkgeneralizationability.Evenifthegeneralizationabilityoftheneuralnetworkweusedisgood,westillgetinaccurateresultssincethenewcomingrulemaynotberelatedtoanyofthepreviousrules.Experimentsshowthisalgorithmissuccessfulinsituationswhichsatisfytheseconditions.

简介：特征关联源于量测过程中的不确定性,是无源多传感器多目标跟踪中一个关键环节。模糊集理论的基本思想是把经典集合中的绝对隶属关系模糊化,为处理不确定事物的建模提供有力工具。一种基于模糊聚类的辐射源特征关联模型被提出,同时给出了确定相似性度量和检验门限的方法。最后模拟产生了雷达数据库,对雷达数据库进行了聚类,并与硬聚类算法和灰色聚类算法进行了比较,实验结果证明了该方法的优势和有效性。

简介：

简介：Ahybridapproachforfuzzysystemdesignbasedonclusteringandakindofneurofuzzynetworksisproposed.Anunsupervisedclusteringtechniqueisfirstlyusedtodeterminethenumberofif-thenfuzzyrulesandgenerateaninitialfuzzyrulebasefromthegiveninput-outputdata.Then,aclassofneurofuzzynetworksisconstructedanditsweightsaretunedsothattheobtainedfuzzyrulebasehasahighaccuracy.Finally,twoexamplesoffunctionapproximationproblemsaregiventoillustratetheeffectivenessoftheproposedapproach.

简介：Dempster-Shaferevidencetheory(DStheory)iswidelyusedinbrainmagneticresonanceimaging(MRI)segmentation,duetoitsefficientcombinationoftheevidencefromdifferentsources.Inthispaper,animprovedMRIsegmentationmethod,whichisbasedonfuzzyc-means(FCM)andDStheory,isproposed.Firstly,theaveragefusionmethodisusedtoreducetheuncertaintyandtheconflictinformationinthepictures.Then,theneighborhoodinformationandthedifferentinfluencesofspatiallocationofneighborhoodpixelsaretakenintoconsiderationtohandlethespatialinformation.Finally,thesegmentationandthesensordatafusionareachievedbyusingtheDStheory.ThesimulatedimagesandtheMRIimagesillustratethatourproposedmethodismoreeffectiveinimagesegmentation.

简介：特征选择(FS)指选择给定的outcome.Unlike是很预兆的那些输入属性的过程另外的维数减小方法，在FS的reduction.The好处是双重的以后，特征选购者保存特征的原来的意思：它更加减少正式就职算法的跑的时间，并且增加在用模糊逻辑和模糊逻辑的乳房X线照片分类的FS过程基于的产生model.This纸分析的精确性Quickreduct算法被申请从特征的FS提取了使用在乳房X线照片region.The上构造的灰色的水平co-occurence矩阵(GLCM)预兆的精确性特征用NaiveBayes，Ripper，C4.5，和蚂蚁矿工algo被测试

简介：Amongtheavailableclusteringalgorithmsindatamining,theCLOPEalgorithmattractsmuchmoreattentionwithitshighspeedandgoodperformance.However,theproperchoiceofsomeparametersintheCLOPEalgorithmdirectlyaffectsthevalidityoftheclusteringresults,whichisstillanopenissue.Forthispurpose,thispaperproposesafuzzyCLOPEalgorithm,andpresentsamethodfortheoptimalparameterchoicebydefiningamodifiedpartitionfuzzydegreeasaclusteringvalidityfunction.TheexperimentalresultswithrealdatasetillustratetheeffectivenessoftheproposedfuzzyCLOPEalgorithmandoptimalparameterchoicemethodbasedonthemodifiedpartitionfuzzydegree.

简介：ThesecondaryusageofspectrumhasbeeninvestigatedinCognitiveRadio(CR)networktoresolvingthespectrumscarcityissueinwirelesscommunication.WhenPrimaryUsers(PU)whoownthespectrumappear,spectrumhandoffisneededtomaintainthecommunicationsofSecondaryUsers.ButthedecisionmakingofspectrumhandoffisachallengeissueforCRnetwork,becausetheinputofdecisionmaking,whichobtainthroughspectrumsensing,isheterogeneousandinexact.Inthispaperwewillusefuzzylogiccontroltheorytosolvethisissueandmakeuseofnewinformationforhandoffoperation:theprobabilityofPU'soccupancyatacertainchannel.Ournewalgorithmcanmakemoreintelligentdecisioncomparedtosimpletraditionalspectrumhandoffdecisionmakingandreducetheprobabilityofspectrumhandoff,alsotheperformanceofSU'scommunicationcanbeenhanced.

简介：古典不平的集合在模糊决定桌子有一个有限处理能力。有古典不平的集合的联合模糊集合，模糊决定桌子上的属性减小算法被学习。首先，新类似度和新类似范畴被定义。同时，被条件属性划分的类似范畴簇当时是provided.And，二条定理被介绍。随后，一个新属性减小算法被建议。最后，新属性减小算法通过自行修整的飞行控制系统的一张表演评估决定桌子被验证。结果证明建议属性减小算法能处理模糊决定桌子到某个程度。

简介：Torealizetheon-linemeasurementandmakeanalysisonthedensityofalgaeandtheirclusterdistribution,thefluorescentdetectionandfuzzypatternrecognitiontechniquesareused.Theprincipleoffluorescentfiber-opticdetectionisgivenaswellasthemethodoffuzzyfeatureextractionusingaclassofneuralnetwork.

简介：ＡＦｕｚｚｙＲｅａｓｏｎｉｎｇＳｙｓｔｅｍａｎｄＩｔｓＨｅｕｒｉｓｔｉｃＩｎｆｅｒｅｎｃｅＡｌｇｏｒｉｔｈｍＺｕｏＸｉａｏｄｅ＆ＬｉａｎｇＹｕｎＤｅｐｔ．ｏｆＢｕｓｉｎｅｓＡｄｍｉｎｉｓｔｒａｔｉｏｎ，ＪｉｎａｎＵｎｉｖｅｒｓｉｔｙ，Ｇｕａｎｇｚｈｏ...

简介：Thispaperproposedafuzzifyfunctorasanextensionoftheconceptoffuzzysets.Thefuzzifyfunctorandthefirst-orderoperatedfuzzysetaredefined.Fromthetheoryanalysis,itcanbeobservedthatwhenthefuzzifyfunctoractsonasimplecrispset,wegetthefirstorderfuzzysetortype-1fuzzyset.Byoperatingthefuzzifyfunctoronfuzzysets,wegetthehigherorderfuzzysetsorhighertypefuzzysetsandtheirmembershipfunctions.Usingthefuzzifyfunctorwecanexactlydescribethetype-1fuzz...

简介：因为表示每份object.In订单解决这个问题使用明确的形式，古典不平的集合不能在决定过程显示出模糊和目标的重要性，这份报纸第一介绍每个目标在有会员度显示出它的模糊的一张特殊决定桌子并且被分配了重量在决定procedure.Then显示出它的重要性，特殊决定桌子被学习，同时，相关不平的集合模型被提供相关definitio

简介：ＦｕｚｚｙＥｎｔｒｏｐｙＢａｓｅｄＣｏｍｂｉｎｅｄＬｅａｒｎｉｎｇＡｌｇｏｒｉｔｈｍｆｏｒＮｅｕｒａｌＮｅｔｗｏｒｋｓ￥ＭｉｎＹａｏ（Ｄｅｐｔ．ｏｆＣｏｍｐｕｔｅｒＳｃｉｅｎｃｅ，ＨａｎｇｚｈｏｕＵｎｉｖｅｒｓｉｔｙ，Ｈａｎｇｚｈｏｕ３１００２８，Ｐ...

简介：Thispaperpresentsaweb-basedsystemtopredicttheelectricityprices.Theproposedsystemcapturesthegeographicallocation,weatherforecast,andoilpriceforoneweekahead.Thecapturedparametersarefedtoafuzzy-logic-basedalgorithmtocalculateelectricenergyprices.Basedonpredictedelectricityprices,consumerscanturnON/OFForrescheduleoperationsoftheirhomeappliancestoreducetheirelectricitybill.Theproposedalgorithmwasdevelopedandhostedinautilityserver(U-server).Ontheconsumerside,ahomegateway(H-gateway),andamonitoringandcontrolsystemwasdesigned,built,andtestedbyusingasinglechipmicrocontroller.