资源描述：
《测量作者研究关联性：基于单词、基于主题和作者共被引方法的比较（英文）》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。
1、MeasuringAuthorResearchRelatedness:AComparisonofWord-Based,Topic-Based,andAuthorCocitationApproachesKunLuandDietmarWolframSchoolofInformationStudies,UniversityofWisconsin-Milwaukee,P.O.Box413,Milwaukee,WI53201.E-mail:{kunlu,dwolfram}@uwm.eduRelationshipsbet
2、weenauthorsbasedoncharacteristicsrelationshipstudiedisbasedonthedataused:directcitation,ofpublishedliteraturehavebeenstudiedfordecades.cocitationanalysis,co-authorshipanalysis,bibliographicAuthorcocitationanalysisusingmappingtechniquescouplinganalysis,andco
3、-wordanalysis(discussedbelow).hasbeenmostfrequentlyusedtostudyhowcloselytwoauthorsarethoughttobeinintellectualspacebasedonAllhavebeensuccessfullyappliedtovisualizescientifichowmembersoftheresearchcommunityco-citetheirstructureandtodescribeauthorrelatedness.R
4、ecently,moreworks.Otherapproachesexisttostudyauthorrelated-sophisticatedhybridmethods(i.e.,usingtextualcontentandnessbasedmoredirectlyonthetextoftheirpublishedcitations)havebeenappliedtothemappingofarticlesworks.Inthisstudywepresentstaticanddynamicword-(Ahl
5、gren&Colliander,2009;Boyack&Klavans,2010;Caobasedapproachesusingvectorspacemodeling,aswellasatopic-basedapproachbasedonlatentDirichletallo-&Gao,2005)andjournals(Liuetal.,2010).Tothebestofourcationformappingauthorresearchrelatedness.Vectorknowledgetheuseofte
6、xtualcontentand,morespecifically,aspacemodelingisusedtodefineanauthorspacecon-topicmodel(e.g.,Deerwester,Dumais,Furnas,Landauer,&sistingofworksbyagivenauthor.OutcomesforHarshman,1990)todeterminetherelatednessofauthorsthetwoword-basedapproachesandatopic-basedh
7、avenotbeenstudiedyet.approachfor50prolificauthorsinlibraryandinforma-tionsciencearecomparedwithmoretraditionalauthorInthisstudyweproposenewtextualfeature-basedcocitationanalysisusingmultidimensionalscalingapproachesbasedonco-occurringwordsthatapplyvectorandh
8、ierarchicalclusteranalysis.Thetwoword-basedspacemodelingtomeasuretherelatednessofauthors’approachesproducedsimilaroutcomesexceptwhereresearch.Atopic-basedapproachusinglatentDirichlettwoauthorswerefrequ
