会员注册 | 登录 | 微信快捷登录 支付宝快捷登录 QQ登录 微博登录 | 帮助中心 人人文库renrendoc.com美如初恋!
站内搜索 百度文库

热门搜索: 直缝焊接机 矿井提升机 循环球式转向器图纸 机器人手爪发展史 管道机器人dwg 动平衡试验台设计

   首页 人人文库网 > 资源分类 > PDF文档下载

外文资料--Broiler Growth Performance Analysis from.PDF

  • 资源星级:
  • 资源大小:260.28KB   全文页数:4页
  • 资源格式: PDF        下载权限:注册会员/VIP会员
您还没有登陆,请先登录。登陆后即可下载此文档。
  合作网站登录: 微信快捷登录 支付宝快捷登录   QQ登录   微博登录
友情提示
2:本站资源不支持迅雷下载,请使用浏览器直接下载(不支持QQ浏览器)
3:本站资源下载后的文档和图纸-无水印,预览文档经过压缩,下载后原文更清晰   

外文资料--Broiler Growth Performance Analysis from.PDF

BroilerGrowthPerformanceAnalysisfromCorrelationAnalysis,MultipleLinearRegression,toNeuralNetworkMeiyanXiao,PeijieHuang,PiyuanLin,ShangweiYanCollegeofInformaticsSouthChinaAgriculturalUniversityGuangzhou,ChinaCorrespondingAuthorpjhuangscau.edu.cnAbstractThepurposeofthisstudyistoinvestigatethedataWeusethebroilergrowthdatasetofthemostfamousfittingforbroilergrowthperformanceparameters.Inthispaper,poultryraisingcompanyinChinatoevaluateourapproachandthegradualadvancinganalysismethods,fromcorrelationtheresultsshowtheeffectivenessofourapproach.analysis,multiplelinearregression,toneuralnetwork,areproposed.Themeantechnologyroadmapisfirstly,correlationTherestofthispaperisorganizedasfollows.Inthenextanalysisisusedtodetectthedegreeofcorrelationbetweenthesection,wepresentthegradualadvancinganalysismethods.broilergrowthperformanceparameterandthecandidateinputExperimentsarepresentedanddiscussedinSection3.Finally,variables.AndthenchoosethepredictorvariablesthathavegoodSection4listssomeconclusions.correlationwiththedependentvariabletobuildthemultiplelinearregressionorneuralnetworkpredictionmodel,orboth,II.GRADUALADVANCINGANALYSISMETHODSaccordingtothelineardegreeofcorrelations.CombinedpredictionmaybechoseoncebothmodelshavegoodpredictionTheexploremethodsinourstudyofdatafittingforbroilerperformances.Weusethebroilergrowthdatasetofthemostgrowthperformanceparametersisdevelopedstepbystep,fromfamouspoultryraisingcompanyinChinatoevaluateourcorrelationanalysistoMLR,andthentononlinearfittingapproachandtheresultsshowtheeffectivenessofourapproach.meansbyneuralnetwork.KeywordsgrowthperformancecorrelationanalysismultipleA.TechnologyRoadmaplinearregressionneuralnetworkbroilerbreedingThetechnologyroadmapofourgradualadvancinganalysismethodsisshowninFig.1.I.INTRODUCTIONBioinformatics1isapromisingyoungfieldthatappliescomputertechnologyinbiologyanddevelopsalgorithmsandmethodstomanageandanalyzebiologicaldata2.Forthemodernpoultrybreedingcompanies,itisdeservedtopredictthepoultrygrowthperformanceparameters,suchasrateforsale,feedintake,dailygainandfeedconversionratio,basedonthemassivehistoricaldatagraduallycumulatedinproduction.However,becauseofthecomplexityanduncertaintybringbytheinfluenceofenvironmentalandphysiologicalfactors,informationintegrationofbiologicaldataisachallenge.Inthispaper,thegradualadvancinganalysismethods,fromcorrelationanalysis,multiplelinearregressionMLR3,toneuralnetwork4,areproposedtostudythedatafittingforbroilergrowthperformanceparameters.Inbroilerbreeding,seasonalfactorplaysanimportantpart.Ontheeffectofseasonalfactors,broilergrowthperformanceFigure1.Technologyroadmapoftheproposedmethodscanbeobviouslydifferent.SobroilergrowthperformanceTheassociationbetweenvariablescanbelinearorparametershaveobviousseasonalvariation.Seasonalfactorsnonlinear.Correlationanalysisismostlyusedtoevaluatelinearincludeairtemperature,precipitation,windspeed,pressure,relationships.Associationsbetweentwovariablescanberelativehumidity,etc.Thispapertakestheinfluenceoftheairanalyzedwithabivariatecorrelationanalysis.Whiletemperaturetotherateforsaleforexampletointroducetheassociationsbetweenonedependentvariableandasetoftwobroilergrowthperformanceanalysismethods.ormoreindependentvariables,whichhavestrongThisworkissupportedbytheSciTechResearchProjectofGuangdongProvinceunderGrantNo.2007A020300010,theNational863HighTechResearchDevelopmentPlanofChinaunderGrantNo.2006AA10Z246,andtheNewDisciplineSupportingFundofSouthChinaAgriculturalUniversityunderGrantNo.2007X022.NotlinearenoughDependentvariableIndependentvariablesCorrelationAnalysisComparisonCombinedpredictionwhenbothhavegoodpredictresponsesMultipleLinearRegressionNeuralNetworkStronglinercorrelation9781424447138/10/25.00©2010IEEEcorrelationswiththedependentvariable,canbestudiedusingmultiplecorrelationregressionanalysis,suchasMLR.Alternatively,ifthedegreeofcorrelationsisnotlinearenoughbetweenthedependentvariableandtheindependentvariables,somenonlinearfittingsprovidegoodchoose.Inthenonlinearfittingmethods,comparingtoGompertzthatusingleastsquaresinnonlinearregression,neuralnetworkisprovedtohasgoodabilitytopredictresponses5.Finally,inpracticalapplication,ifbothMLRandneuralnetworkhavegoodpredictionperformances,wecanconsiderthecombinedprediction.B.CorrelationAnalysisAcorrelationanalysisisastatisticalprocedurethatevaluatestheassociationbetweenthedependentvariableandtheindependentvariablesrespectively.Thesimplestwaytofindoutqualitativelythecorrelationistoplotthedata.AndwecanquantifythedegreeofcorrelationbyspecifyingthecorrelationcoefficientR,definedasyyinixxiyxnRσµσµ−−−∑1111wherexµandxσdenotethesamplemeanandthesamplestandarddeviationrespectivelyforthevariablexandyµandyσdenotethesamplemeanandthesamplestandarddeviationrespectivelyforthevariabley.Assumethataperfectlinearrelationshipexistsbetweenthevariablesxandy,i.e.,baxyiifori1,2,...,nwith0≠a.Nowverifyusingthedefinitionsofthemeanandthevariancethatbaxyµµandxyaσσ.Thisimpliesfrom1thatRa/|a|.Orinotherwords,R1ifa0andR1ifa0.ThecaseR1correspondstothemaximumpossiblelinearpositiveassociationbetweenxandy,meaningthatallthedatapointswilllieexactlyonastraightlineofpositiveslope.Similarly,R1correspondstothemaximumpossiblenegativeassociationbetweenthestatisticalvariablesxandy.Ingeneral,1≤R≤1withthemagnitudeandthesignofRrepresentingthestrengthanddirectionrespectivelyoftheassociationbetweenthetwovariables.C.MultipleLinearRegressionOncewehaveestablishedthatastrongcorrelationexistsbetweenthedependentvariableandmorethanoneindependentvariable,wewilluseMLR.AlinearregressionmodelthatcontainsmorethanonepredictorvariableiscalledaMLRmodel.ThefollowingmodelisaMLRmodelwithtwopredictorvariables,1xand2xuxxy2210βββ2Themodelislinearbecauseitislinearintheparameters,0β,1βand2β.Themodeldescribesaplaneinthethreedimensionalspaceofy,1xand2x.Theparameter0βistheinterceptofthisplane.Parameters1βand2βarereferredtoaspartialregressioncoefficients.Parameter1βrepresentsthechangeinthemeanresponsecorrespondingtoaunitchangein1xwhen2xisheldconstant.Parameter2βrepresentsthechangeinthemeanresponsecorrespondingtoaunitchangein2xwhen1xisheldconstant.uistherandomerror.D.NeuralNetworkNeuralnetwork4offeranalternativetoregressionanalysisforbiologicalmodeling.Inrelationtosystemmodeling,thedifferencebetweenartificialneuralnetworksandregressionanalysisisthatanequationisnotassumed,tighterfitsofdataarepossible,anditispossibletoworkwithnoisydata.Verylittleresearchhasbeenconductedtomodelanimalgrowthusingartificialneuralnetworks5,6.Inourstudy,wechoosetheBackPropagationBPneuralnetwork,whichisafeedforwardmultilayernetworkbasedontheBackPropagationalgorithmdevelopedbyRumelhartandMcCelland7andhasbecomeoneofthemostwidelyusedneuralnetworkinpractice.TheActivationTransferFunctionATFofaBPnetwork,usually,isadifferentiableSigmoidSshapefunction,whichhelpstoapplynonlinearmappingfrominputstooutputs.AtwolayerBPnetworkwasusedinourmodel.ThegoodnessoffitsfortheobtainedneuralnetworkmodelwascalculatedbymeansquareerrorMSEandmeanpercentageerrorMPE.TheMPEandMSEarecomputedas∑−nttttyyynMPE1ˆ13nyyMSEnttt∑−12ˆ4wheretyequalstheobservedvalueattimet,tyˆequalstheestimatedvalue,andnequalsthenumberofobservations.III.EXPERIMENTALRESULTSA.ExperimentSetupWetakethebreedingareaofGuangdongprovinceofChinaforexampletoevaluateourapproach.ThedatasetoftendaymeanairtemperatureisprovidedbyGuangdongProvincialClimateandAgrometeorologicalCente.AndthebroilergrowthdatasetisprovidedbyGuangdongWensFoodGroupLimitedCompany,whichisthemostfamouspoultryraisingcompanyinChina.AndwetakehenofshortfeetbuffBforexampletoevaluatetheinfluenceoftheairtemperaturetotherateforsale.Weselecthengrowthdataof2007,whichconsistsof5714data,andremain4209dataafterdatapreprocessing,whichistoeliminateabnormaldata,suchasabnormalrateforsale,nulldayage,andnullweight.FortheMLRandneuralnetworkmodels,weselect70samplesrandomlyfortraining,andtherestfortesting.B.CorrelationAnalysisConsideringthatthefullgrowingstageofbroilercanbedividedintochicklingstagethefirst4weeksandadultchickenstage.Differentstageshavedifferentphysiologicalcharacteristic.So,inourstudy,firstly,weusescatterplotstoshowtherelationshipbetweentherateforsaleandthetendaymeanairtemperatureofhen,chicklingstage,andadultchickenstagerespectively,whichareshowninFig.2toFig.4.Andthen,thedegreesofcorrelationsarequantifiedbycorrelationcoefficientR,whichisshowninTable1.0.920.930.940.950.960.970.9871217222732Tendaymeanairtemperature℃RateforsaleFigure2.TendaymeanairtemperatureofhenVSrateforsale0.940.950.960.970.9871217222732Tendaymeanairtemperature℃RateforsaleFigure3.TendaymeanairtemperatureofchicklingstageVSrateforsale0.940.950.960.970.9871217222732Tendaymeanairtemperature℃RateforsaleFigure4.TendaymeanairtemperatureofadultchickenstageVSrateforsaleTABLEI.CORRELATIONCOEFFICIENTCaseRTendaymeanairtemperatureofhenandrateforsale0.8506Tendaymeanairtemperatureofchicklingstageandrateforsale0.8932Tendaymeanairtemperatureofadultchickenstageandrateforsale0.8594AswecanseefromTable1,correlationcoefficientRoftherateforsaleandthetendaymeanairtemperatureofchicklingstageandadultchickenstageisbiggerthanthatoftherateforsaleandthetendaymeanairtemperatureofhen,whichindicatesthedivisionofchicklingstageandadultchickenstagetodofurtherresearchisarightchoose.C.MultipleLinearRegressionThefollowingMLRequationisfitforthetrainingdata21057.00755.0367.93xxy5whereyistherateforsale,and1xand2xarethetendaymeanairtemperatureofchicklingstageandadultchickenstagerespectively.D.NeuralNetworkSimilartotheMLRmodel,weusethetendaymeanairtemperatureofchicklingstageandadultchickenstageasinputs,andsettherateforsaleasoutput.Fig.5showstherealobservedvaluesandpredictedrateforsaleforbothMLRandneuralnetworklabeledasNNinFig.5methods,usingthetestingdata.Figure5.ComparsionofMLRandneuralnetworkinpredictionTable2showsthestatisticsfortheMLRandneuralnetworkforpredictingbroilerrateforsale.TABLEII.MODELSTATISTICSFORMLRANDNEURALNETWORKFORPREDICTINGRATEFORSALEModelStatisticMPEMSEMLR0.524.328E05Neuralnetwork0.473.538E05AswecanseefromTable2,neuralnetworkmodeloutperformsMLRmodelinbothMPEandMSE.Butfromtheresult,wecanseetheMLRmodelalsohasgoodpredictionperformances.IV.CONCLUSIONSInthispaper,wehavedealtwiththeresearchofthedatafittingforbroilergrowthperformanceparameters.Gradualadvancinganalysismethods,fromcorrelationanalysis,MLR,toneuralnetwork,areproposed.WeusethebroilergrowthdatasetofthemostfamouspoultryraisingcompanyinChina,andtakestheinfluenceoftheairtemperaturetotherateforsaleforexampletoevaluateourapproach.Aswecanseefromexperiment,correlationanalysisisusedtodetectthatthedivisionofchicklingstageandadultchickenstageisgoodforfurtherresearch,sincethetendaymeanairtemperatureofthesetwostageshavebiggercorrelationcoefficientRwithrateforsalethanthatofthetendaymeanairtemperatureofhen.Neuralnetworkmodelhasbetterabilitytopredictresponses.ButwecanseetheMLRmodelalsohasgoodpredictionperformance.So,wecanconcludethattheMLRandneuralnetworkmodelsbuiltbythetendaymeanairtemperatureofchicklingstageandadultchickenstagebothhavegoodpredictionperformancesandaresuitforcombinedpredictionforrateforsaleinpracticalapplication.REFERENCES1J.Cohen,BioinformaticsAnintroductionforcomputerscientists,ACMComputingSurveys,362,122158,2004.2J.HanandM.Kamber,DataMiningConceptsandTechniques,2ndedition,MorganKaufmann,2006.3S.Weisberg,AppliedLinearRegression,3rdedition.NewYorkWiley,2005.4M.T.Hagan,H.B.Demuth,M.H.Beale,Neuralnetworkdesign,PWSPublishedcompany,1996.5W.B.Roush,W.A.DozierIII,S.L.Branton,Comparisonofgompertzandneuralnetworkmodelsofbroilerchickens,PoultryScience.85794797,2006.6D.Yee,M.G.Prior,andL.Z.Florence,Developmentofpredictivemodelsoflaboratoryanimalgrowthusingartificialneuralnetworks,Comput.Appl.Biosci.9517–522,1993.7D.E.Rumelhart,J.L.McCelland,Learningrepresentationsbybackpropagatingerrors,Nature,3236188533536,1986.

注意事项

本文(外文资料--Broiler Growth Performance Analysis from.PDF)为本站会员(图纸帝国)主动上传,人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。 若此文所含内容侵犯了您的版权或隐私,请立即通知人人文库网([email protected]),我们立即给予删除!

温馨提示:如果因为网速或其他原因下载失败请重新下载,重复下载不扣分。

copyright@ 2015-2017 人人文库网网站版权所有
苏ICP备12009002号-5