资源描述：
《BIOINFORMATICS ORIGINAL PAPER》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、Vol.21no.22005,pages248–256BIOINFORMATICSORIGINALPAPERdoi:10.1093/bioinformatics/bth496GenenameambiguityofeukaryoticnomenclaturesLifengChen1,∗,HongfangLiu2andCarolFriedman11DepartmentofBioMedicalInformatics,ColumbiaUniversity,NewYork,NY10032,USAand2Departmen
2、tofInformationSystems,UniversityofMaryland,BaltimoreCounty,Baltimore,MD21250,USAReceivedonApril11,2004;revisedonAugust19,2004;acceptedonAugust20,2004AdvanceAccesspublicationAugust27,2004ABSTRACTnamesthatwerealsoEnglishwordswereincluded,233%Motivation:Withmor
3、eandmorescientificliteraturepublishedadditional‘gene’instanceswereretrieved,mostofwhichwereonline,theeffectivemanagementandreuseofthisknow-falsepositives.Wealsofoundthatauthorsprefertousesyn-ledgehasbecomeproblematic.Naturallanguageprocessingonyms(74.7%)tooffi
4、cialsymbols(17.7%)orfullnames(7.6%)(NLP)maybeapotentialsolutionbyextracting,structur-intheirpublications.ingandorganizingbiomedicalinformationinonlineliteratureContact:lifeng.chen@dbmi.columbia.eduinatimelymanner.Oneessentialtaskistorecognizeandidentifygenom
5、icentitiesintext.‘Recognition’canbeaccom-INTRODUCTIONplishedusingpatternmatchingandmachinelearning.ButforOwingtotheadvancementofhigh-throughputtechnologies,‘identification’thesetechniquesarenotadequate.Inordersuchasgenomicsequencing,DNAchipsandpowerfulcom-toi
6、dentifygenomicentities,NLPneedsacomprehensiveputers,theamountofscientificliteraturehasbeenincreasingresourcethatspecifiesandclassifiesgenomicentitiesasexponentiallyinthepastdecade.Muchoftheliteraturehastheyoccurintextandthatassociatesthemwithnormalizedbeenpubli
7、shedonlineaswellasinjournalarticles.Thetermsandalsouniqueidentifierssothattheextractedentitiesplethoraoftheonlineliteratureisaninvaluableresourcearewelldefined.Onlineorganismdatabasesareanexcel-forresearchersifusedeffectively.However,theinforma-lentresourcetoc
8、reatesuchalexicalresource.However,tionisstoredinheterogeneousformatsbydifferentgroupsofgenenameambiguityisaseriousproblembecauseitaffectsresearchersandusedforvariouspurposes.Thismakesman-theappr
