DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS

DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS

ID:38237294

大小:143.85 KB

页数:5页

时间:2019-05-27

DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS_第1页
DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS_第2页
DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS_第3页
DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS_第4页
DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS_第5页
资源描述:

《DATA SIGNATURES FOR VALIDATION AND EVALUATION OF TEMPORAL ASSOCIATIONS》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库

1、DATASIGNATURESFORVALIDATIONANDEVALUATIONOFTEMPORALASSOCIATIONSGeorgeK.GeorgiouBonK.Sy12QueensCollegeandTheGraduateSchoolandUniversityCenteroftheCityUniversityofNewYork/DepartmentofComputerScience165-30KissenaBoulevard,Flushing,NewYork,11367-0904233West42ndStreet,NewYork,NY

2、10036fgeorgiou;bong@qcvaxa:acc:qc:eduAbstractthecomputationalcostinvolvedinsuchatask,weproposeatwostepapproachthatactsnotonlyasaheuristicforDiscoveringrelationsamongdomainvariablesfromdataimprovedeciencyinoursearch,butalsoasa ltertotrapplaysanimportantroleinautomatingand/o

3、rassistingthederivationofcoincidentaltemporalassociations.Firstly,processofconstructingdomainmodelsandvalidatingex-wepreprocessthetime-dependentdatasetofeachvariable,istingones.Animportantkindofrelationisthetemporalextractingonlyitsuniquepatterns.Datathatappearinassociation

4、betweendomainvariables.Whileastraightfor-achronologicalsequenceisreferredtoastime-dependentwardapplicationofcorrelationanalysismaybeinsucientdata;e.g.,meteorologicaldatasuchasthetemperatureofforuncoveringtheserelationships,weproposeanapproachadayinNewYorkCity,or nancialdat

5、asuchasthedailythatattemptstoidentifysimilaritiesinthepatternsacrosstradingvolumeorthedailyclosingpriceofastock.Eachdatasets.Thesepatternsenableustocapturethetempo-uniquepatterniscalledadatasignature.Adatasignatureralassociationsamongthevariablesofamodel.isablockofconsecuti

6、vetimevaryingdatapointswhichexhibitslowornocorrelationtoanyotherblockofdata1INTRODUCTIONinthesamedataset.Forinstance,inFig.1atherearetwoModelingtemporalassociationsamongasetofvariablessignatureswhereasinFig.1bthereisone.requiresanexplicitrepresentationofbothcovariationandte

7、mporalprecedenceamongthevariables.Buildingsuchadynamicmodelismoresuitablethanastaticmodelfortherepresentationofadynamicsystem.Nevertheless,thelevelofdicultyinvolvedintheconstructionofadynamicmodelincreasesdramaticallyovertheconstructionofastaticmodelduetoanadditionaldimens

8、ion,time.In-corporatingthisadditionaldimensioninamodelrequiresthetaskofidentifying

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。