简介:
简介:Thispaperdealswithanimportantsubjectofrejectingtheabnormaldataintelligentlyinthedynamicdatasystem.Basedontheprincipleofnearestneighboroffuz-zymathematics,anapproachofmathematicallyabstractingthehumanthinkingandphys-icalpracticeknowledgeisdiscussed,anewmethodofautomaticrejectionofabnormalda-taisthenproposed.Theexperimentalresultsshowthatthemethodisavailabletothepractice.
简介:数据存取延期成为了高端计算系统的突出的性能瓶颈。在系统设计减少数据存取延期的关键是减少数据货摊时间。存储器地区和并发是影响现代存储器系统的性能的二个必要因素。因为全面存储器系统性能上的存储器并发的影响很好没被理解,然而,存在在利用数据存取并发上在很少减少数据货摊时间学习焦点。在这研究,一双新奇数据货摊时间模型,为地区和并发的联合努力的L-C模型和为数据上的纯失误的效果的下午模型阻止时间,被介绍。模型提供数据存取延期的新理解并且为表演优化提供新方向。基于这些新模型,先进缓存优化的一张概括表格被介绍。当时,被数据并发贡献了,把38个条目仅仅,21个条目由数据地区作出贡献,它显示出数据并发的值。在这研究介绍的L-C和下午模型和他们的联系结果和机会为数据中央的建筑学和算法现代计算系统设计的未来重要、必要。
简介:Inthisagecharacterizedbyrapidgrowthinthevolumeofdata,datadeidentificationtechnologieshavebecomecrucialinfacilitatingtheanalysisofsensitiveinformation.Forinstance,healthcareinformationmustbeprocessedthroughdeidentificationproceduresbeforebeingpassedtodataanalysisagenciesinordertopreventanyexposureofpersonaldetailsthatwouldviolateprivacy.Assuch,privacyprotectionissuesassociatedwiththereleaseofdataanddatamininghavebecomeapopularfieldofstudyinthedomainofbigdata.Asastrictandverifiabledefinitionofprivacy,differentialprivacyhasattractednoteworthyattentionandwidespreadresearchinrecentyears.Inthisstudy,weanalyzetheadvantagesofdifferentialprivacyprotectionmechanismsincomparisontotraditionaldeidentificationdataprotectionmethods.Furthermore,weexamineandanalyzethebasictheoriesofdifferentialprivacyandrelevantstudiesregardingdatareleaseanddatamining.
简介:Thisarticlepresentsadatamanagementsolutionbasedonthedatadistributionservice(DDS)communicationmodel.ThebasicDDScommunicationmodelconsistsofaunidirectionaldataexchangewhereapplicationsthatpublishdata'push'therelevantdata,whichisupdatedtothelocalcachesofco-locatedsubscriberstothedata[1].DDShasnospecifiedcenternodetoforwarddatapacketsandmaintainthecommunicationdata.Thistypeofpublish-subscribe(P/S)modelpresentsintegrityandconsistencychallengesindatamanagement.Unlikepeer-to-peer(P2P)distributedstorage,DDSapplicationshaveahardreal-timeenvironmentandfewerdatafeatures,andthecoreproblemisensuringtheintegrityandconsistencyofdataindistributedsystemsunderthishardreal-timeenvironment.ThisarticlebeginswithabriefintroductionofthecommunicationmodelusedbyDDS,thenanalyzespersistentdatamanagementproblemscausedbysuchmodel,andprovidesanappropriatesolutiontotheseproblems.Thissolutionhasbeenimplementedinaprototypesystemofthereal-timeservicebus(RTSB)ofTsinghuaUniversity.
简介:Theapplicationofdataenvelopmentanalysis(DEA)asamultiplecriteriadecisionmaking(MCDM)techniquehasbeengainingmoreandmoreattentioninrecentresearch.InthepracticeofapplyingDEAapproach,theappearanceofuncertaintiesoninputandoutputdataofdecisionmakingunit(DMU)mightmakethenominalsolutioninfeasibleandleadtotheefficiencyscoresmeaninglessfrompracticalview.ThispaperanalyzestheimpactofdatauncertaintyontheevaluationresultsofDEA,andproposesseveralrobustDEAmodelsbasedontheadaptationofrecentlydevelopedrobustoptimizationapproaches,whichwouldbeimmuneagainstinputandoutputdatauncertainties.TherobustDEAmodelsdevelopedarebasedoninput-orientedandoutputorientedCCRmodel,respectively,whentheuncertaintiesappearinoutputdataandinputdataseparately.Furthermore,therobustDEAmodelscoulddealwithrandomsymmetricuncertaintyandunknown-but-boundeduncertainty,inbothofwhichthedistributionsoftherandomdataentriesarepermittedtobeunknown.TherobustDEAmodelsareimplementedinanumericalexampleandtheefficiencyscoresandrankingsofthesemodelsarecompared.TheresultsindicatethattherobustDEAapproachcouldbeamorereliablemethodforefficiencyevaluationandrankinginMCDMproblems.
简介:TheLHCexperimentsatCERNwillgeneratehugevolumesofdata-severalPBperyearatdataratesbetween100MB/sand1.5GB/s.Thestorageandanalysisofthesedatapresentamajorchallenge.IncollaborationwithothermembersoftheformerRD45project,thecentraldatabasesupportgroupatCERNhasbeenworkingonthisissueforseveralyears,leadingtoproductionuseofapotentialsolution,basedonthecombinationofanObjectDatabaseandMassStoragesystem,botheatCERNandoutside.
简介:Throughanalyzingtheprincipleofdatasharinginthedata-basesystem,thispaperdiscussestheprincipleandmethodforintegratingandsharingGISdatabydataengine,introducesawaytoachievethehighintegrationandsharingofGISdataonthebasisofVCTinVC++,andpro-videsthemethodforunitingVCTintoRDBMSinordertoimplementaspa-tialdatabasewithobject-orienteddatamodel.
简介:Thefastturnoverofsoftwaretechnologies,inparticularinthedomainofinteractivity(coveringuserinterfaceandvisualisation)makesitdifficultforasmallgroupofpeopletoproducecompleteandpolishedsoftware-toolsbeforetheunderlyingtechnologiesmakethemobsolete.AttheHepVis'99workshop,aworkinggrouphasbeenformedtoimprovetherpoductionofsoftwaretoolsfordataanalysisinHENP.Besidepromotingadistributeddevelopmentorganisation,onegoalofthegroupistosystematicallydesignasetofabstractinterfacesbasedonusingmodernOOanalysisandOOdesigntechniques.Aninitialdomainanalysishascomeupwithseveralcategories(componets)foundintypicaldataanalysistools:historams,Ntuples,Functions,Vectors,Fitter,Plotter,AnalyzerandController,SpecialEmphasiswasputonreducingthecouplingsbetweenthecategoriestoaminimum,thusoptimisingre-useandmaintainabilityofanycomponentindividually.TheinterfaceshavebeendefinedinJavaandC++andimplementationsexistintheformoflibrariesandtoolsusingC++(Anaphe/Lizard,Openscientist)andJava(JavaAnalysisStudio),AspecialimplementationaimsataccessingtheJavaLiraries(throughtheirAbstractInterfaces)fromC++.ThispapergiveranoverviewofthearchitectureanddesignofthevariouscomponentsfordataanalysisasdiscussedinAIDA.
简介:Since1998,theALICEexperimentandtheCERN/ITdivisionhavejointlyexecutedseverallarge-scalehighthroughputdistributedcomputingexercises:theALICEdatachallenges.ThegoalsoftheseregularexercisesaretotesthardwareandsoftwarecomponentsofthedataacqusitionandcomputingsystemsinrealisticconditionsandtoexecuteanearlyintegrationoftheoverallALICEcomputinginfrastructure.ThispaperreportsonthethirdALICEDataChallenge(ADCIII)thathasbeenperformedatCERNfromJanuarytoMarch2001.ThedatausedduringtheADCⅢaresimulatedphysicsrawdataoftheALICETPC,producedwiththeALICEsimulationprogramAliRoot.ThedataacquisitionwasbasedontheALICEonlineframeworkcalledtheALICEDataAcquisitionTestEnvironment(DATE)system.Thedataaftereventbuilding,werethenformattedwiththeROOTI/OpackageandadatacataloguebasedonMySQlwasestablished.TheMassStorageSystemusedduringADCIIIisCASTOR.Differentsoftwaretoolshavebeenusedtomonitortheperformances,DATEhasdemonstratedperformancesofmorethan500MByte/s.Anaggregatedatathroughputof85MByte/swassutainedinCASTORoverseveraldays.Thetotalcollecteddataamountsto100TBytesin100.00files.
简介:Multisensordatafusion(MDF)isanemergingtechnologytofusedatafrommultiplesensorsinordertomakeamoreaccurateestimationoftheenvironmentthroughmeasurementanddetection.ApplicationsofMDFcrossawidespectruminmilitaryandcivilianareas.Withtherapidevolutionofcomputersandtheproliferationofmicro-mechanical/electricalsystemssensors,theutilizationofMDFisbeingpopularizedinresearchandapplications.ThispaperfocusesonapplicationofMDFforhighqualitydataanalysisandprocessinginmeasurementandinstrumentation.Apractical,generaldatafusionschemewasestablishedonthebasisoffeatureextractionandmergeofdatafrommultiplesensors.Thisschemeintegratesartificialneuralnetworksforhighperformancepatternrecognition.AnumberofsuccessfulapplicationsinareasofNDI(Non-DestructiveInspection)corrosiondetection,foodqualityandsafetycharacterization,andprecisionagriculturearedescribedanddiscussedinordertomotivatenewapplicationsintheseorotherareas.ThispapergivesanoverallpictureofusingtheMDFmethodtoincreasetheaccuracyofdataanalysisandprocessinginmeasurementandinstrumentationindifferentareasofapplications.
简介:AMethodofImprovingSeismicDataResolution:ComprehensiveInversionofWellloggingandSeismicDataZhangYufen;HongFeng(DepartmentofAppl...
简介:IthaslongbeenacknowledgedthatGISdatacanbeusedasauxiliaryinformationtoimproveremotesensingimageclassification.Inpreviousstudies,GISdatawereoftenusedintrainingareaselectionandpostprocessingofclassificationresultoractedasadditionalbands.Generally,itisfulfilledinastatisticalorinteractivemanner,soitisdifficulttousetheauxiliarydataautomaticallyandintelligently. Furthermore,iftheclassifierrequestscertainstatisticalcharacteristics,theadditionalbandmethodcannotbeusedbecausemostauxiliarydatadonotmeettherequirementsofstatisticalcharacteristics.Ontheotherhand,expertsystemtechniqueswereincorporatedinremotesensingimageclassificationtomakeuseofdomainknowledgeandlogicalreasoning.Butbuildinganimageclassificationexpertsystemwasverydifficultbecauseofthe“knowledgeacquisitionbottleneck”. Spatialdataminingandknowledgediscovery(SDMKD),istheextractionofimplicit,interestingspatialornon_spatialpatternsandgeneralcharacteristics.Weproposedatheoreticalandtechnicalframeworkofspatialdataminingandknowledgediscovery(Lietal.,1997).Andspatialdataminingissupposedtobeusedintwoaspects,oneisintelligentanalysisofGISdata,theotheristosupportknowledgedriveninterpretationandanalysisofremotesensingimages.SDMKDprovidesanewwayofknowledgeacquisitionforremotesensingimageclassification.Severalresearchershavedonesomeworkinthisfield.Eklundetal.(1998)extractedknowledgefromTMimagesandgeographicdatainsoilsalinityanalysisusinginductivelearningalgorithmC4.5.Huangetal.(1997)extractedknowledgefromGISdataandSPOTmultispectralimageinwetlandclassificationusingC4.5too.Inthesetwostudies,geographicdatawereconvertedfromvectortorasterformatinwhichthesamplingsizeisequaltoimagepixelsize.Theimplementationofdataminingtechniquesinspatialdatabase,especiallyinductivelearningmethod,andthecombinationo