9.REFERENCES
[1]S.L.Abebe,V.Arnaoudova,P.Tonella,G.Antoniol,and
Y.Gueheneuc.Canlexiconbadsmellsimprovefaultprediction?InWCRE,2012.
[2]M.Allamanis,E.T.Barr,C.Bird,andC.Sutton.Learning
naturalcodingconventions.InFSE,2014.
[3]M.AllamanisandC.Sutton.Miningsourcecoderepositories
atmassivescaleusinglanguagemodeling.InMSR.IEEEPress,2013.
[4]M.AllamanisandC.Sutton.MiningIdiomsfromSource
Code.InFSE,2014.
[5]V.Arnaoudova,M.DiPenta,G.Antoniol,andY.-G.
Gueheneuc.Anewfamilyofsoftwareanti-patterns:Linguisticanti-patterns.InCSMR,2013.
[6]V.Arnaoudova,L.M.Eshkevari,M.D.Penta,R.Oliveto,
G.Antoniol,andY.Guéhéneuc.REPENT:analyzingthenatureofidenti errenamings.IEEETSE,2014.
[7]V.Arnaoudova,M.D.Penta,andG.Antoniol.Linguistic
antipatterns:Whattheyareandhowdevelopersperceivethem.EMSE,2015.
[8]M.Banko,V.Mittal,andM.Witbrock.Headlinegeneration
basedonstatisticaltranslation.InACL,2000.
[9]K.Beck.Implementationpatterns.PearsonEducation,2007.[10]Y.Bengio,R.Ducharme,P.Vincent,andC.Janvin.Aneural
probabilisticlanguagemodel.TheJournalofMachineLearningResearch,2003.
[11]D.Binkley,M.Hearn,http://www.77cn.com.cnwrie.Improvingidenti er
informativenessusingpartofspeechinformation.InMSR,2011.
[12]http://www.77cn.com.cnpositionalmorphologyforword
representationsandlanguagemodelling.InICML,2014.[13]S.Butler,M.Wermelinger,Y.Yu,andH.Sharp.Relating
identi ernaming awsandcodequality:Anempiricalstudy.InWCRE,2009.
[14]S.Butler,M.Wermelinger,Y.Yu,andH.Sharp.Exploringthe
in uenceofidenti ernamesoncodequality:Anempiricalstudy.In14thEuropeanConferenceonSoftwareMaintenanceandReengineering(CSMR’2010,pages156–165,2010.
[15]S.Butler,M.Wermelinger,Y.Yu,andH.Sharp.MiningJava
classnamingconventions.InICSM,2011.
[16]S.ChenandJ.Goodman.Anempiricalstudyofsmoothing
techniquesforlanguagemodeling.InACL,1996.
[17]T.A.Corbi.Programunderstanding:Challengeforthe1990s.
IBMSystemsJournal,28(2):294–306,1989.
[18]A.DeLucia,M.DiPenta,R.Oliveto,A.Panichella,and
http://www.77cn.com.cningIRmethodsforlabelingsourcecodeartifacts:Isitworthwhile?InICPC,2012.
[19]B.Dorr,D.Zajic,andR.Schwartz.Hedgetrimmer:A
parse-and-trimapproachtoheadlinegeneration.InHLT-NAACL-03,2003.
[20]B.P.Eddy,J.A.Robinson,N.A.Kraft,andJ.C.Carver.
Evaluatingsourcecodesummarizationtechniques:Replicationandexpansion.InICPC,2013.
[21]M.U.GutmannandA.Hyvärinen.Noise-contrastive
estimationofunnormalizedstatisticalmodels,withapplicationstonaturalimagestatistics.TheJournalofMachineLearningResearch,2012.
[22]S.Haiduc,J.Aponte,andA.Marcus.Supportingprogram
comprehensionwithsourcecodesummarization.InICSE,2010.
[23]S.Haiduc,J.Aponte,L.Moreno,andA.Marcus.Ontheuse
ofautomatedtextsummarizationtechniquesforsummarizingsourcecode.InWCRE,2010.
[24]D.Hendrix,J.Cross,S.Maghsoodloo,etal.Theeffectiveness
ofcontrolstructurediagramsinsourcecodecomprehensionactivities.IEEETSE,2002.
[25]A.Hindle,E.T.Barr,Z.Su,M.Gabel,andP.Devanbu.On
thenaturalnessofsoftware.InICSE,2012.
[26]E.W.HøstandB.M.Østvold.Debuggingmethodnames.In
ECOOP,2009.
[27]S.Karaivanov,V.Raychev,andM.T.Vechev.Phrase-based
statisticaltranslationofprogramminglanguages.InOnward!,2014.
[28]R.Kiros,R.Zemel,andR.Salakhutdinov.Multimodalneural
languagemodels.InNIPS,2013.
[29]http://www.77cn.com.cnwrie,C.Morrell,H.Feild,andD.Binkley.What’sina
name?astudyofidenti ers.InICPC,2006.
[30]B.Liblit,A.Begel,andE.Sweetser.Cognitiveperspectiveson
theroleofnamingincomputerprograms.InProceedingsofthe18thAnnualPsychologyofProgrammingWorkshop,2006.[31]C.J.MaddisonandD.Tarlow.Structuredgenerativemodels
ofnaturalsourcecode.InICML,2014.
[32]C.D.Manning,P.Raghavan,andH.Schütze.Introductionto
InformationRetrieval.CambridgeUniversityPress,2008.[33]R.C.Martin.Cleancode:ahandbookofagilesoftware
craftsmanship.PearsonEducation,2008.
[34]S.McConnell.CodeComplete.MicrosoftPress,2004.[35]T.Mikolov,K.Chen,G.Corrado,andJ.Dean.Ef cient
estimationofwordrepresentationsinvectorspace.InICLRWorkshop,2013.
[36]T.Mikolov,W.-t.Yih,andG.Zweig.Linguisticregularitiesin
continuousspacewordrepresentations.InHLT-NAACL,2013.[37]A.MnihandG.Hinton.Threenewgraphicalmodelsfor
statisticallanguagemodelling.InICML,2007.
[38]A.MnihandK.Kavukcuoglu.Learningwordembeddings
ef cientlywithnoise-contrastiveestimation.InNIPS,2013.[39]A.MnihandY.W.Teh.Afastandsimplealgorithmfor
trainingneuralprobabilisticlanguagemodels.InICML,2012.[40]L.Mou,G.Li,Z.Jin,L.Zhang,andT.Wang.TBCNN:a
tree-basedconvolutionalneuralnetworkforprogramminglanguageprocessing.arXivpreprintarXiv:1409.5718,2014.[41]T.T.Nguyen,A.T.Nguyen,H.A.Nguyen,andT.N.Nguyen.
Astatisticalsemanticlanguagemodelforsourcecode.InFSE,2013.
[42]J.Pennington,R.Socher,andC.D.Manning.Glove:Global
vectorsforwordrepresentation.EMNLP,2014.
[43]V.Raychev,M.Vechev,andA.Krause.Predictingprogram
propertiesfrom“bigcode”.InPOPL,2015.
[44]S.RussellandP.Norvig.Arti cialIntelligence:AModern
Approach.PrenticeHall,1995.
[45]G.Sridhara.Automaticgenerationofdescriptivesummary
commentsformethodsinobject-orientedprograms.UniversityofDelaware,2012.
[46]G.Sridhara,E.Hill,D.Muppaneni,L.Pollock,and
K.Vijay-Shanker.Towardsautomaticallygeneratingsummarycommentsforjavamethods.InASE,2010.
[47]G.Sridhara,L.Pollock,andK.Vijay-Shanker.Automatically
detectinganddescribinghighlevelactionswithinmethods.InICSE,2011.
[48]N.Srivastava,G.Hinton,A.Krizhevsky,I.Sutskever,and
R.Salakhutdinov.Dropout:Asimplewaytopreventneuralnetworksfromover tting.JournalofMachineLearningResearch,2014.
[49]A.Takang,P.Grubb,andR.Macredie.Theeffectsof
commentsandidenti ernamesonprogramcomprehensibility:anexperientialstudy.JournalofProgramLanguages,