数据挖掘算法在交通状态量化及识别的应用

©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.:2007-11-30:(50738004):(1979-),,,,:;(1976-),,,,:;(1983-),,,,::1001-9081(2008)03-0738-041,2,1(1.,201804;2.,201804)([e-mail]):(ITS),,,,,:;;;:TP311.13:AApplicationofdataminingintrafficstatequantificationandrecognitionSUNYa1,QIANHong2bo2,YELiang1(1.SchoolofTransportationEngineering,TongjiUniversity,Shanghai201804,China;2.ControlTheoryandControlEngineeringMobileStationforPostdoctors,TongjiUniversity,Shanghai201804,China)Abstract:IntheIntelligentTransportationSystem(ITS)environment,themagnanimousflowinformationoftrafficdetectorwastargeted,usingdataminingtechnologywhichincludeddatacollection,datapreprocessing,datamining,resultanalysisandappraisalandpatternapplicationtocarryonnewinformationextraction.Variousstagesrequestaswellastheclusteringanalysisandthepatternrecognitionalgorithmwereproposed.Finallythenewusefulinformationtrafficstatewasobtainedfromthemagnanimousdata.Simultaneouslythereal2timegatheringtrafficdatawasusedtodistinguishthetrafficstate,whichfinallyindicatedtherecognitionstatecouldaccuratelyreflecttheactualtrafficstate.Keywords:collectedinformationfromdetector;datamining;trafficstate;clusteringanalysis2070,McMasterMacroscopic,,,,,,1,1(IntelligentTransportationSystem,ITS),ITS,,,,,ITS[1],,,:,,,,122.1K2[2],,,28320083ComputerApplicationsVol.28No.3Mar.2008©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(FuzzyC2Means,FCM),FCM:1)U0,2)U0(1)Vi3)(2)Vi,U,4),,,UVi2)4),:Jm(U,V)=ni=1cj=1umijxi-vj2(1):cj=1uij=1;uij0,cj1,ni1(2):n,{,,}cuijij,jiVxi-yijim,xiFCM,,c:V=[v1,v2,,vc]=v11v21vc1v12v22vc2v13v23vc3(3)V:vi1,pcu/h;vi2,km/h;vi3,:i=1,2,,cV:v1=[v11,v12,v13]T,,;v2=[v21,v22,v23]T,,;vc=[vc1,vc2,vc3]T,c,c2.2[3],V,V=[v1,v2,,vc],V,x:xi-vj=sk=1(xik-vjk)2(4)k=1,2,,c;xi;vj,:xi-vjh=minj{xi-vj}(5)xivjk,,c,,,,3,,(SQL)Oracle,Excel,1(SeviceID)A20SVD20001,1A20SVD20001FDT_RECORDTIMEFSTR_SDEVICEIDFSTR_LOOPSTATUS1FINT_LANEFLOW1FINT_LANESPEED1FINT_LANERATIO1*04-03-209:01A20SVD20001257712*04-03-209:02A20SVD2000117858*04-03-209:03A20SVD20001267912********3.1,3.1.1[4],,,,:1),,,2),,(,9373:©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.),[5]:1)0,0,2)0,0,3)[0,100],4),5)120km/h,1800pcu/h,,()3.1.2,:,,(),()(),,:1),,DATAIDID2)(pcu/h)3.1.3,,,:,:[6],():[80,100][70,80][60,70][50,60][40,50][30,40][20,30][10,20][0,10];3.1.4{,,},,,,,[-1,1][0,1]:1),,:c(i)=(c(i)-min(c(i)))/(max(c(i)-min(c(i))))(6)2)c,mean(v)sd(v),c,(7)c(i)=(c(i)-mean(v))/sd(v)(7)3.2,Matlab[7],,:V=[v1,v2,v3]=8402160113058.329.416.76.333.457.5(8)V1,pcu/h;2,km/h;3,23.3,3[8],1[840,58.3,6.3],820pcu/h,6.3%,58.3km/h,,3,,[840,58.3,6.3],3,2[2160,29.4,33.4],,3[1130,16.7,57.5],3.44(a)IDA20SVD10087,4(b),,1237:007:52,,,404728©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(b),,7:5211:15,A20SVD10087,,,,,4(b),4(a),,44,,,,:[1]HANDD,MANNILAH,SMYTHP.Principlesofdatamining[M].Cambridge:MassachusettsLondonEngland,TheMITPress,2001:6-8.[2]KIRSCHFINKH.Basictoolsforfuzzymodeling[C/OL].[2007-09-15].[3],.[M].:,1992:169-178.[4].[D].:,2006:2-8.[5]CHENCHAO.Detectingerrorsandimputingmissingdataforsingleloopsurveillancesystems[C]//Proceedingsof82ndTransportationResearchBoard(TRB)AnnualMeeting.Washington,DC:IEEEPress,2003:6-16.[6]ISHAKS.Quantifyinguncertaintiesoffreewaydetectorobservationsusingfuzzy2clusteringapproach[C]//TransportationResearchRe2cord,TransportationResearchBoard82ndAnanualMeeting.Wash2ington,DC:IEEE,2003:2-7.[7].Matlab[M].:,2004.[8].[M].:,1987:22-24.(728),ID40Eid=2,:{c,b,d,g,h,a,a,b,f,e,g,a,c,b,g}:{2.0,2.0,1.8,1.6,1.1,1.0,0.9,0.9,0.8,0.7,0.6},e,g1,0.60,0.6,,,143,,,,,,,,:[1]AGRAWALR,IMIELINSKIT,SWAMIA.Miningassociationrulesbetweensetsofitemsinlargedatabases[C]//ProceedingsoftheACMSIGMODConferenceonManagementofData.Washington,DC:ACMPress,1993:207-216.[2]AGRAWALR,SRIKANTR.Fastalgorithmforminingassociationrules[C]//ProceedingsofInternationalConferenceonVeryLargeDataBases.Santiago:IEEEPress,1994:487-499.[3]PARKJS,CHENMINI2SYAN,YUPS.Efficientparalleldataminingforassociationrules[C]//Proceedingsof4thInternationalConferenceonKnowledgeandDataEngineering.NewYork:ACMPress:1995:31-36.[4]AGRAWALR,SHAFERJ.Parallelminingofassociationrules[J].IEEETransonKnowledgeandDataEngineering,1996,8(6):962-969.[5]ZAKIM.SPADE:Anefficientalgorithmforminingfrequentse2quences[J].MachineLearning,2001,41(1/2):31-60.[6]PEIJ,HANJ,PINTOH,etal.PrefixSpan:Miningsequentialpat2ternsefficientlybyprefix2projectedpatterngrowth[C]//ProceedingsofIntConferenceonDataEngineering(ICDE01).Heidelberg:IEEEPress,2001:215-224.[7]WILLER.RestructuringLatticetheory:Anapproachbasedonhier2archiesofconcepts[J].OrderedSets,1982,11(5):445-470.[8],.[J].,2004,31(10A).[9]YUNU.Anewframeworkfordetectingweightedsequentialpatternsinlargesequencedatabases[J/OL].[2007-04-19].: