Query vector space model VSM design and active node selection.docx

Query vector space model VSM design and active node selection.docx

《Query vector space model VSM design and active node selection.docx》由会员分享，可在线阅读，更多相关《Query vector space model VSM design and active node selection.docx（24页珍藏版）》请在冰豆网上搜索。

QueryvectorspacemodelVSMdesignandactivenodeselection

Sulcalandgyrallandmarksonthehumancerebralcortexarerequiredforvariousstudiesofthehumanbrain.Whetheruseddirectlytoexaminesulcalgeometry,orindirectlytodrivecorticalsurfaceregistrationmethods,theaccuracyoftheselandmarksisessential.Whileseveralmethodshavebeendevelopedtoautomaticallyidentifysulciandgyri,theiraccuracymaybeinsufficientforcertainneuroanatomicalstudies.Wedescribeasemi-automatedprocedurethatdelineatesasulcusorgyrusgivenalimitednumberofuser-selectedpoints.Themethodusesagraphtheoryapproachtoidentifythelowest-costpathbetweenthepoints,wherethecostisacombinationoflocalcurvaturefeaturesandthedistancebetweenverticesonthesurfacerepresentation.Weimplementedthealgorithminaninterfacethatguidestheuserthroughacorticalsurfacedelineationprotocol,andweincorporatedthistoolintoourBrainSuitesoftware.WeperformedastudytocomparetheresultsproducedusingourmethodwithresultsproducedusingDisplay,apopulartoolthathasbeenusedextensivelyformanualdelineationofsulcallandmarks.Sixratersweretrainedonthedelineationprotocol.Theyperformeddelineationson12brainsusingbothsoftwarepackages.Weperformedastatisticalanalysisof3aspectsofthedelineationtask:

timerequiredtodelineatethesurface,registrationaccuracyachievedcomparedtoanexpert-delineatedgold-standard,andvariationamongraters.Ournewmethodwasshowntobefastertouse,toprovidereducedinter-ratervariability,andtoprovideresultsthatwereatleastasaccurateasthoseproducedusingDisplay.

ArticleOutline

1.Introduction

2.Methods

2.1.Curve-trackingprocedure

2.1.1.Convexitymeasure

2.1.2.Graphweighting

2.1.3.Pathcomputation

2.2.Softwareimplementation

2.2.1.Landmarkidentification

2.2.2.Protocolinterface

2.3.Validationmethodology

2.3.1.Timeefficiency

2.3.2.Delineationaccuracy

2.3.3.Inter-ratervariability

3.Results

3.1.Timeperformance

3.2.Delineationaccuracy

3.3.Inter-ratervariability

4.Discussion

Acknowledgements

References

Purchase

$35.95

386

Areviewontheintegrationofartificialintelligenceintocoastalmodeling  OriginalResearchArticle

JournalofEnvironmentalManagement,Volume80,Issue1,July2006,Pages47-57

KwokwingChau

 Closepreview  |  Relatedarticles  |  Relatedreferenceworkarticles    

AbstractAbstract|Figures/TablesFigures/Tables|ReferencesReferences

Abstract

Withthedevelopmentofcomputingtechnology,mechanisticmodelsareoftenemployedtosimulateprocessesincoastalenvironments.However,thesepredictivetoolsareinevitablyhighlyspecialized,involvingcertainassumptionsand/orlimitations,andcanbemanipulatedonlybyexperiencedengineerswhohaveathoroughunderstandingoftheunderlyingtheories.Thisresultsinsignificantconstraintsontheirmanipulationaswellaslargegapsinunderstandingandexpectationsbetweenthedevelopersandpractitionersofamodel.Therecentadvancementsinartificialintelligence（AI）technologiesaremakingitpossibletointegratemachinelearningcapabilitiesintonumericalmodelingsystemsinordertobridgethegapsandlessenthedemandsonhumanexperts.Theobjectiveofthispaperistoreviewthestate-of-the-artintheintegrationofdifferentAItechnologiesintocoastalmodeling.Thealgorithmsandmethodsstudiedincludeknowledge-basedsystems,geneticalgorithms,artificialneuralnetworks,andfuzzyinferencesystems.Morefocusisgiventoknowledge-basedsystems,whichhaveapparentadvantagesovertheothersinallowingmoretransparenttransfersofknowledgeintheuseofmodelsandinfurnishingtheintelligentmanipulationofcalibrationparameters.Ofcourse,theotherAImethodsalsohavetheirindividualcontributionstowardsaccurateandreliablepredictionsofcoastalprocesses.Theintegratedmodelmightbeverypowerful,sincetheadvantagesofeachtechniquecanbecombined.

ArticleOutline

1.Introduction

2.Numericalmodeling

2.1.Modelmanipulation

2.2.Generationsofmodeling

3.IncorporationofAIintomodeling

3.1.Knowledge-basedsystems（KBSs）

3.2.Geneticalgorithms（GAs）

3.3.Artificialneuralnetworks（ANNs）

3.4.Fuzzyinferencesystems

4.Futuredirections

5.Conclusions

Acknowledgements

References

Purchase

$41.95

387

DevelopmentofaCAD/CAMsystemfortheclosed-dieforgingprocess  OriginalResearchArticle

JournalofMaterialsProcessingTechnology,Volume138,Issues1-3,20July2003,Pages436-442

M.Jolgaf,A.M.S.Hamouda,S.Sulaiman,M.M.Hamdan

 Closepreview  |  Relatedarticles  |  Relatedreferenceworkarticles    

AbstractAbstract|Figures/TablesFigures/Tables|ReferencesReferences

Abstract

Theapplicationofcomputer-aidedengineering（CAE）,computer-aideddesign（CAD）,andcomputer-aidedmanufacturing（CAM）,isessentialinmodernmetalformingtechnology.Thus,theprocessofmodellingfortheinvestigationandunderstandingofdeformationmechanicshasbecomeamajorconcerninrecentandadvancedresearch,andthefiniteelementmethod（FEM）hasassumedincreasedimportance,particularlyinthemodellingofdeformationprocesses.ThisworkisdevotedtothedevelopmentofaCAD/CAMsystemfortheclosed-dieforgingprocess.Thesystemdevelopmentconsistsofthreestages:

namely,metalflowsimulation,diefailureanalysisanddesignoptimisation,andthedevelopmentandimplementationofamachiningcode.Inthefirststage,theFEMwasusedtosimulatetheaxisymmetricclosed-dieforgingprocessofcoppermaterial.Themethodisusedtostudythemetalflow;diefilling,retainingthenon-linearityinvolvedinthelargechangeinthegeometry;thecontinuouschangeinthecontactsurfacecondition;andtheisotropicmaterialwork-hardeningcharacteristics.Inthesecondstage,afiniteelementanalysisandoptimisationalgorithmisdevelopedtoexaminethediefatiguelifeandtooptimisethediedesign.ThefiniteelementanalysisinthefirstandsecondstagewascarriedoutusingcommerciallyavailablefiniteelementsoftwarecalledLUSAS.Inthethirdstage,amachiningcodefortheoptimiseddieisdevelopedandimplementedusingCAD/CAMsoftwarecalledUniGraphicsandCNCmachine.

ArticleOutline

1.Introduction

2.Designingthediecavity

3.FEsimulationofaxisymmetricclosed-dieforging

3.1.Thesimulationmodel

3.2.Materialproperties

3.3.Geometry

3.4.Theboundaryconditions

3.5.Meshconvergencestudy

4.Resultsanddiscussion

4.1.Load–displacementcurve

4.2.Materialflowbehaviour

4.3.Displacementvectors

4.4.Stressdistribution

5.Dieanalysisandoptimisation

5.1.Diestressesanddisplacementdistribution

5.2.Dielifeandfatigueanalysis

5.3.CAMforthedie

6.Conclusions

References

Purchase

$37.95

388

Foodserviceresearch:

Anintegratedapproach  ReviewArticle

InternationalJournalofHospitalityManagement,Volume30,Issue2,June2011,Pages477-483

SvetlanaRodgers

 Closepreview  |  Relatedarticles  |  Relatedreferenceworkarticles    

AbstractAbstract|Figures/TablesFigures/Tables|ReferencesReferences

Abstract

Foodproductiondemandscompetenciesinbothnaturalscienceandmanagementprinciples.Thesuggestedframeworkincludesthefollowingareasofresearch:

technologicalinnovation,processdesign（includingIndustrialCuisine）,productdevelopment（includingMolecularGastronomy,IndustrialGastronomyandFunctionalMeal）andriskmanagement（includingquantitativeriskassessmentandpredictivemicrobiology）.Developmentsstemmingfromfoodscienceandengineeringcouldbecombinedwithstudiesinoperationsmanagement,consumerpsychology,cognitivepsychology,sociologyandeconomics.Thepaperconcludesthatinordertomeetsocietaldemandsforhealth,sustainabilityandefficiencyaradicalshiftisneeded—from‘lowtech’recipedevelopmentandproliferationofmarketingconceptsto‘cuttingedge’technologicalinnovationsupportedbyadvancedresearchinmanagementandconsumerstudies.

ArticleOutline

1.Introduction

2.Definingresearchonfoodproduction

3.Technologicalinnovation

4.Processdesign

5.Productdevelopment

6.Riskmanagement

7.Conclusion

References

Purchase

$41.95

389

EnhancingVoIPserviceforubiquitouscommunicationinacampusWLANwithpartialcoverage  OriginalResearchArticle

ComputerNetworks,Volume52,Issue13,17September2008,Pages2489-2504

Hung-YunHsieh,You-EnLin,Hsiao-PuLin

 Closepreview  |  Relatedarticles  |  Relatedreferenceworkarticles    

AbstractAbstract|Figures/TablesFigures/Tables|ReferencesReferences

Abstract

TheIEEE802.11WLANtechnologyhasbecomethedefactostandardforwirelessInternetaccess.ThespottycoverageofWLANaccesspoints,however,confinestheapplicabilityofmanyreal-timeservicessuchasVoIPwithintheboundaryoftheWLANservicearea.Inthispaper,weinvestigatetheproblemofenhancingVoIPserviceforubiquitouscommunicationinaWLANwithspottyservicearea.WeconsiderauniversitycampusthathasanestablishedinfrastructureforsupportingSIP-basedVoIPservicethrougheitherwiredorwirelessdatanetworks.ThecampusWLANservicedoesnothave100%fullcoverage,andhenceuserscannotmakeuntetheredVoIPcallsanywhereoncampus.Thegoalofthispaperistoovercomethelimitationsofsuch“deadspots”formotivatingtheuseofcampusIPtelephonyservice.Toproceed,westartwithtwoapproachescalledone-hopextensionanddual-modecommunication.Thefirstapproachusesmulti-hoprelaytoextendtheWLANcoverage,whilethesecondapproach