1 / 10
文档名称:

Analysis of International Competitiveness of China’s Mobile Phone Industry Based on Data Mining Algorithm 2022 Jun Lu.pdf

格式:pdf   大小:878KB   页数:10页
下载后只包含 1 个 PDF 格式的文档,没有任何的图纸或源代码,查看文件列表

如果您已付费下载过本站文档,您可以点这里二次下载

Analysis of International Competitiveness of China’s Mobile Phone Industry Based on Data Mining Algorithm 2022 Jun Lu.pdf

上传人:紫鹃 2023/1/12 文件大小:878 KB

下载得到文件列表

Analysis of International Competitiveness of China’s Mobile Phone Industry Based on Data Mining Algorithm 2022 Jun Lu.pdf

相关文档

文档介绍

文档介绍:该【Analysis of International Competitiveness of China’s Mobile Phone Industry Based on Data Mining Algorithm 2022 Jun Lu 】是由【紫鹃】上传分享,文档一共【10】页,该文档可以免费在线阅读,需要了解更多关于【Analysis of International Competitiveness of China’s Mobile Phone Industry Based on Data Mining Algorithm 2022 Jun Lu 】的内容,可以使用淘豆网的站内搜索功能,选择自己适合的文档,以下文字是截取该文章内的部分文字,如需要获得完整电子版,请下载此文档到您的设备,方便您编辑和打印。Hindawi
SecurityandCommunicationNetworks
Volume2022,ArticleID8460574,10pages
/2022/8460574
ResearchArticle
AnalysisofInternationalCompetitivenessofChina’sMobile
PhoneIndustryBasedonDataMiningAlgorithm
JunLuandZhongLiLv
SchoolofEconomicsandManagement,GuizhouNormalUniversity,Guiyang550001,China
CorrespondenceshouldbeaddressedtoJunLu;******@
Received17February2022;Revised16March2022;Accepted21March2022;Published23April2022
AcademicEditor:Chin-LingChen
Copyright©2022JunLuandZhongLiLv.,isisanopenaccessarticledistributedundertheCreativeCommonsAttribution
License,whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkis
properlycited.
InordertoanalyzetheinternationalcompetitivenessofChina-mademobilephones,thispapercombinesdataminingalgorithms
toanalyzetheinternationalcompetitivenessofChina’smobilephoneindustry,improvetheinternationalmarketshareofChina-
mademobilephones,,thepapercarries
outthemodeldescriptionoftheselectedresearchmethodscombiningprincipalcomponentanalysis,dataenvelopmentanalysis,
,thispaperexpoundsontheprinciplesandmodelsofPCA,DEA,
,thispaperverifiesthevalidityofthemodel
proposedinthispaper,conductsstatisticalanalysisthroughthedataminingmodel,evaluatesthedataminingeffectthrough
multiplesimulationexercises,andverifiesthevalidityofthesystemmodel.

valueisanimportantresearchidea.
Facedwiththeproblemsofhighmarketshareandlowprofit,isarticlecombinesdataminingalgorithmstoanalyze
marginsofdomesticmobilephonecompanies,clusteringoftheinternationalcompetitivenessofChina’smobilephone
low-endandmiddle-endmarkets,weaknessinthehigh-endindustry,increasethemarketshareofdomesticmobile
market,andlowprofits,inthefiercemarketcompetition,phonesintheworld,andprovideareferenceforthesub-
timelyadjustmentofbusinessstrategiestoenhancethesequentdevelopmentofourcountry’smobilephone
competitivenessofdomesticmobilephonecompanieshasindustry.
,asconsumersof
corporateproductsandservices,
customerscanbesaidtobeanewandimportantsourceof
corporatecompetitiveness.,erefore,itisnecessaryto,etheoryofenterprisecompetitivenessoriginatesfromthe
-researchontheacquisitionandmaintenanceofcompeti-
-
product,,itcanbedividedintotwomajortheoreticalschools:
ordertogainstrongcompetitiveness,domesticmobileexogenoustheoryandendogenoustheory[1].,eschoolof
phonecompaniesshouldestablishacustomer-orientedexogenoustheoryfocusesontheanalysisoftheexternal
businessphilosophy,enhancecustomerperceivedvalue,andcompetitiveenvironmentofenterprisesandbelievesthatthe
regularlymonitorandutilizecustomerperceivedvalue,soascompetitivenessofenterprisesisreflectedinthecomparison
[2]pointsoutthatenterprises
serviceswithhighercustomervalue.,erefore,thinkingdesign,manufacture,andsellgoodsorprovideservicesin

2SecurityandCommunicationNetworks
competitors,theabilityandopportunitythatismoreat-competitivenessthroughempiricalresearchandprovided
tractiveinpriceandqualityisthecompetitivenessofen-newresearchideasandresearchmethodsforthestudyof
terprises;literature[3][12]
competitors,theabilityofenterprisestoacquire,create,andrefinesthetechnologicalcompetitivenessofenterprises

-anddevelopment,technologyintegration,andtechnology
tributedthedominantpositionofenterprisesintheindustrymonopolyintheprocessofindependentinnovationand
[4]pointedestablishesastructuralequationmodelbasedonthepath
outthatthecoreofenterprisecompetitivenessiscompar-relationshipbetweenstrategicorientation,technological
ativeproductivity,whichitselfisbasedontheinterindi-competitiveness,[13]
vidual,-
abovescholarsdefinethecompetitivenessofenterprises,petitivenessbasedonthe“push-pullmodel”and“human
althoughtheyconsiderdifferentangles,theyallfocusonthetechnologysymbiosismodel.”Literature[14]usesthe
analysisoftheexternalcompetitiveenvironmentofenter-AHP,basedonthepatent-basedLEDenterprisetech-
prises,whichbelongstotheschoolofexogenoustheory.,enologycompetitivenessevaluationsystem,analyzesthe
endogenousschoolfocusesontheanalysisoftheinternalinnovationinputandoutputdataofLEDpackaginglisted
[5]believesthattheenterprises,andevaluatestheenterprise’stechnological
[15]usestherelativetechnical
thecompetitivenessofthesustainableandlong-termde-advantagesofpatentstodefinethetechnicalstrengthof
[6]attributestheeachresearchobjectanddrawsapatentportfoliochart
competitivenessoftheenterprisetotheinterioroftheen-withthetechnicalattractivenessofeachcompany’s
-mainstreamtechnology,therelativepositionofpatents,
ture[7][16]em-

Intheexplorationoftheconnotationoftechnologicalemotionalvalue.
competitivenessofenterprisesfromdifferentperspectives,
theviewsofvariousscholarshavenotbeenunified,
canbebasicallydividedintotwoschools:theabilityschool

studies,theliterature[8]integratedtheviewpointsoftheIndataanalysis,weoftenfacetheproblemsofjudgingthe
capabilityschoolandtheresourceschool,,the
enterpriseitselfisacombinationofresourcesandcapa-factorsthataffectthecharacteristicsofacertainthingandits
bilities,andpointedoutthatthetechnologicalcompeti-
tivenessofenterprisesstemsfromthefulluseofinternaldeeply,weneedtoanalyzeandevaluatevariousinfluencing
andexternaltechnologicalinnovationresources,,multivariateandlargesample
thefulluseofinternalandexternaltechnologicalinno-datawillbringaboutproblemssuchasmulticollinearity,and
vationresources.,eroleoftheunique,scarce,andir-theduplicationofinformationreflectedbyeachinfluencing
replaceabletechnologicalassetsownedbyanenterprisefactorwillaffecttheauthenticityandscientificityofthe
enablesitselftoprovidemoreattractiveproductsorser-statisticalresults.,erefore,inordertoavoidinformation
vicesinthemarketandtoobtainmorelong-termbenefitsoverlapandreduceworkloadasmuchaspossible,peopleput
“dimensionalityreduction.”Principal
competitivenessofanenterprise,itcanpromotetheeffi-componentanalysisisperformedbyfindingafewuncor-
cientandrationaluseofallitsownresourcesandmaximizerelatedvariables,calculatingtheirlinearcombination,and
,
importanceoftechnologyinthedevelopmentofenter-thesametime,itsavesmostoftheinformationinthe
prises,especiallyhigh-techenterprises,hasbecomein-originalvariabledata,whichisthemostwidelyusedmul-
.
defeatitscompetitors,itmustimproveitstechnological,eprincipalcomponentsareseveralcomprehensive
competitiveness[9].indexesformedbytheoriginalindexesthroughaseriesof
Literature[10]comprehensivelyusestheanalytichi-
mathematicalcalculations,thatis,F1,F2,andF3inthe
erarchyprocess,theentropyweightcoefficientmethod,
Delphimethod,andtheregressionanalysismethodtocontainedintheprincipalcomponents,theyarecalled“first
evaluatethetechnologicalcompetitivenessofourcoun-principalcomponent,”“secondprincipalcomponent,”
try’sself-ownedbrandautomobileenterpriseswiththe“thirdprincipalcomponent,”andsoon.
technologicalinputandoutputcapabilitiesofenterprises
[11]verifiedthe
.
ofcrossanalysisandthecorrespondingcross-analysisX1,X2,X3,...,Xpisap-dimensionalrandomvariable,and
methodfortheevaluationofenterprisetechnologicalprincipalcomponentanalysisisdonetotransformp
SecurityandCommunicationNetworks3
observedvariablesintopnewindicatorsthroughlinearminimizesinputwhileensuringmaximumoutput.,eDEA
combination,namely,methodisanonparametricestimationmethod,whichis
suitablefordealingwithproblemsthatrequiredecision-
FμX+μX+···+μX,
11111221ppmakingorevaluationindicators,andthegoalitselfdoesnot
F2μ21X1+μ22X2+···+μ2pXp,,theDEA
(1)method,basedonlinearprogrammingtheory,performsa
M,seriesofmathematicalcalculationsoninputindexand
Fpμp1X1+μp2X2+···+
evaluatestherelativeeffectivenessofdecision-makingcases
,emodelmeetsthefollowingconditions:withthesameattributes.
,eDEAmethodmainlyevaluatestherelativeefficiency
(1)μ2+μ2+···+μ21
i1i2ipofthedecision-makingunit(DMU).EachDMUhasthe
(2)Cov(Fi,Fj)0,i≠j,i,j1,2,...,psameinputvariablesandoutputvariables.,eDEAmethod
(3)Var(F1)≫Var(F2)≫≫Var(FP)obtainsthecomprehensiveefficiencyindexofeachDMUby
calculatingtheweightedratioofoutputandinputdataand
,epurposeofprincipalcomponentanalysisisto
rankstheDMUsaccordingtothisvaluetodeterminethe
simplifyvariables,andthenumberofprincipalcompo-
effectiveDMU,whichisthedecision-makingunitwiththe
nentsisusuallylessthanthenumberoforiginalvariables.
highestrelativeefficiency.
However,asfortheactualproblems,severalprincipal
Next,wewillintroducetheclassicDEAmodel:theC2R

-makingunits
theprincipalcomponentsreflectasmuchinformationof
andeachdecision-makingunithasmkindsofinputvari-
theoriginalvariablesaspossible.,erefore,itisnecessary
ablesandskindsofoutputvariables,which,respectively,
toweighthenumberofprincipalcomponentsandthe
representthe“resourcesconsumed”and“effectsofwork”of
retainedinformation.,e“information”hereismea-
thedecision-makingunit,inFigure1.
suredbyvariance;thatis,thelargerVar(F)is,themore
1Now,wewanttoevaluatetheefficiencyofthei-th
,thenewindicators0
1decision-,
F,F,F,...F(k≤p)fullyretainthemaininformation
123kDMUisabbreviatedasDMU,(X,Y)as(X,Y),andh
intheoriginalvariablesandareindependentofeachi0ii006
ash.,eclassicC2Rmodeloptimizestheweightcoeffi-

cientsuandvtomaximizeh0,whichisusedtosolvethe
followingoptimizationproblem:
.,especificstepsof
sy
principalcomponentanalysisareasfollows:r1μrr0
maxh0n,
jvjxj
(1),ealgorithmcalculatesthecorrelationcoefficient10

r1rri,i,...,n,
(2),ealgorithmcalculatesthecharacteristicrootofthen≤11()
jvjxji2
correlationcoefficientmatrixandthecorresponding1
characteristicvector.
ur>0,r1,...,s,
(3),ealgorithmselectsthelargestfeatureroot,andthe
correspondingfeaturevectorisequaltothecoeffi-vj>0,j1,...,m.
,the
algorithmselectsthesecondlargestfeatureroot,andHere,y