download0 view153
twitter facebook

공공누리This item is licensed Korea Open Government License

Title
Feasibility Classification of New Design Points Using Support Vector Machine Trained by Reduced Dataset
Author(s)
정승현정민중
Publication Year
2012-05-01
Abstract
Inthispaper,weproposetouseasupportvectormachine(SVM)fortheclassificationofdesigndata.AlthoughtheSVMisaverypopulartechniqueindatamining,itisrarelyappliedtoanindustrialdesignprocessthatmayrequireinformationregardingthefeasibilityofthedesignpointofinterest.Tocheckthefeasibility,thedesignermustconductexperimentsorcomputersimulations,whichmayincurconsiderablecost.Therefore,theSVMcanbeaneffectivetoolforpredictingfeasibleandinfeasibleregionsbecauseitonlyusesthecumulativedesigndata.Inthispaper,weusedtheSVMtoclassifysampledatasetsdrawnfrommathematicaltestproblemsandfromanair-conditionerpipedesignexample.OurresultsindicatethattheSVMiscapableofveryaccuratelyidentifyingfeasibleandinfeasibleregionsinthedesignspace.Further,
wewereabletoreducethetrainingtimeoftheSVMbyusingthek-meansclusteringalgorithmtoreducetheamountoftrainingdata,takingadvantageofthepowerfulgeneralizationabilitiesoftheSVM.Consequently,weconcludethattheSVM
canbeaneffectivetooltoassessfeasibilityatcertaindesignpoints,avoidingsomeofthehighcomputationalcostsoftheanalysis.
Keyword
Supportvectormachine(SVM); Feasibilityclassification; K-meansclustering; Air-conditionerpipedesignproblem
Journal Title
Internationaljournalofprecisionengineeringandmanufacturing
Citation Volume
13
ISSN
2234-7593
Files in This Item:
There are no files associated with this item.
Appears in Collections:
7. KISTI 연구성과 > 학술지 발표논문
URI
https://repository.kisti.re.kr/handle/10580/14080
http://www.ndsl.kr/ndsl/search/detail/article/articleSearchResultDetail.do?cn=JAKO201217355624195
Export
RIS (EndNote)
XLS (Excel)
XML

Browse