国外应用文和学术论文的格式.docx
- 文档编号:10291575
- 上传时间:2023-05-24
- 格式:DOCX
- 页数:12
- 大小:27.14KB
国外应用文和学术论文的格式.docx
《国外应用文和学术论文的格式.docx》由会员分享,可在线阅读,更多相关《国外应用文和学术论文的格式.docx(12页珍藏版)》请在冰点文库上搜索。
国外应用文和学术论文的格式
Applicationpaper
Thepurposeoftheapplicationpaperistogiveyouanopportunitytoapplytheconceptsofpsychologytoyoureverydaylife.Theapplicationpaperrequiresyoutointegrateatheoryfromtheinformationwe'vestudiedwithaneventoreventsfromyourpersonalexperience.Inotherwords,youaretouseatheorytoanalyzeandtoexplainsomethingthathastakenplaceinyourlife.Thisisnotatheoreticalreviewpapernorisitasimplereportofapersonalexperience.Anin-depthreviewofthetheoreticalmaterialthatyouaregoingtodiscussisnotnecessary.ThisissimplyrepeatingwhatIcanfindinourtextorinclassnotes.Itwouldbeunnecessaryworkforyoutousespacetoprovideanin-depthreviewofmaterialalreadyavailable.Instead,yourtaskbecomesoneofusingthisknowledgetoconsiderviableexplanationsforyourand/orothers'behavior.Youwanttouseyourexampleasanillustrationofthetheoryyou'vechosen,withyourgoalbeingtoconvincemethatyouactuallydounderstandwhatthetheoryandconceptsareallabout.
*Technicalpresentationcovers:
*PaperTitle
*TheAbstract
*TheIntroduction
*RelatedWork
*TheBody
*PerformanceExperiments
*TheConclusions
*FutureWork
*TheAcknowledgements
*Citations
*Appendices
*GrammarandSmall-ScalePresentationIssues
*Mechanics
*VersionsandDistribution
TipsforWritingTechnicalPapers
JenniferWidom,January2006
∙HerearethenotesfromapresentationIgaveattheStanfordInfoLabFridaylunch,1/27/06,withafew(notmany)revisionswhenIreprisedthetalkon12/4/09.Thepresentationcovered:
PaperTitle
∙TheAbstract
∙TheIntroduction
∙RelatedWork
∙TheBody
∙PerformanceExperiments
∙TheConclusions
∙FutureWork
∙TheAcknowledgements
∙Citations
∙Appendices
∙GrammarandSmall-ScalePresentationIssues
∙Mechanics
∙VersionsandDistribution
RunningExample
Asarunning(fictitious!
)example,supposeyou'vedesignedandrunexperimentswithanewalgorithmforexternalmultipassmerge-sort.YouralgorithmreducesthecomplexityfromO(nlogn)toO(n),underthepremisethatit'sacceptabletohavesomebounded"unsortedness"intheresult.Youplantowriteuptheresultsforsubmissiontoamajorconference.
Note:
ThisexamplewasusedthroughoutthelivepresentationbutIhaven'tfollowedthroughmuchinthesenotes.Thus,thenotesincludeseveralexercisesforthereader.
PaperTitle
∙Titlescanbelonganddescriptive:
Linear-TimeExternalMultipassSortingwithApproximationGuarantees
orshortandsweet:
∙ApproximateExternalSort
Here'samiddle-of-the-roadlength,plusacutenamethatsticksinpeople'sminds:
∙Floosh:
ALinear-TimeAlgorithmforApproximateExternalSort
TheAbstract
Statetheproblem,yourapproachandsolution,andthemaincontributionsofthepaper.Includelittleifanybackgroundandmotivation.Befactualbutcomprehensive.Thematerialintheabstractshouldnotberepeatedlaterwordforwordinthepaper.
(Exercise:
Writeanabstractforthemultiwaysortexample.)
TheIntroduction
TheIntroductioniscruciallyimportant.BythetimearefereehasfinishedtheIntroduction,he'sprobablymadeaninitialdecisionaboutwhethertoacceptorrejectthepaper--he'llreadtherestofthepaperlookingforevidencetosupporthisdecision.AcasualreaderwillcontinueoniftheIntroductioncaptivatedhim,andwillsetthepaperasideotherwise.Again,theIntroductioniscruciallyimportant.HereistheStanfordInfoLab'spatentedfive-pointstructureforIntroductions.Unlessthere'sagoodargumentagainstit,theIntroductionshouldconsistoffiveparagraphsansweringthefollowingfivequestions:
1Whatistheproblem?
2Whyisitinterestingandimportant?
3Whyisithard?
(E.g.,whydonaiveapproachesfail?
)
4Whyhasn'titbeensolvedbefore?
(Or,what'swrongwithpreviousproposedsolutions?
Howdoesminediffer?
)
5Whatarethekeycomponentsofmyapproachandresults?
Alsoincludeanyspecificlimitations.
(Exercise:
Answerthesequestionsforthemultiwaysortexample.)
Thenhaveafinalparagraphorsubsection:
"SummaryofContributions".Itshouldlistthemajorcontributionsinbulletform,mentioninginwhichsectionstheycanbefound.Thismaterialdoublesasanoutlineoftherestofthepaper,savingspaceandeliminatingredundancy.
(Exercise:
Writethebulletlistforthemultiwaysortexample.)
RelatedWork
∙Theperennialquestion:
Shouldrelatedworkbecoverednearthebeginningofthepaperorneartheend?
Beginning,ifitcanbeshortyetdetailedenough,orifit'scriticaltotakeastrongdefensivestanceaboutpreviousworkrightaway.InthiscaseRelatedWorkcanbeeitherasubsectionattheendoftheIntroduction,oritsownSection2.
∙End,ifitcanbesummarizedquicklyearlyon(intheIntroductionorPreliminaries),orifsufficientcomparisonsrequirethetechnicalcontentofthepaper.InthiscaseRelatedWorkshouldappearjustbeforetheConclusions,possiblyinamoregeneralsection"DiscussionandRelatedWork".
TheBody
Guideline#1:
Aclearnewimportanttechnicalcontributionshouldhavebeenarticulatedbythetimethereaderfinishespage3(i.e.,aquarterofthewaythroughthepaper).
Guideline#2:
Everysectionofthepapershouldtellastory.(Don't,however,fallintothecommontrapoftellingtheentirestoryofhowyouarrivedatyourresults.Justtellthestoryoftheresultsthemselves.)Thestoryshouldbelinear,keepingthereaderengagedateverystepandlookingforwardtothenextstep.Thereshouldbenosignificantinterruptions--thosecangointheAppendix;seebelow.
Asidefromtheseguidelines,whichapplytoeverypaper,thestructureofthebodyvariesalotdependingoncontent.Importantcomponentsare:
∙RunningExample:
Whenpossible,usearunningexamplethroughoutthepaper.ItcanbeintroducedeitherasasubsectionattheendoftheIntroduction,oritsownSection2or3(dependingonRelatedWork).
∙Preliminaries:
Thissection,whichfollowstheIntroductionandpossiblyRelatedWorkand/orRunningExample,setsupnotationandterminologythatisnotpartofthetechnicalcontribution.Oneimportantfunctionofthissectionistodelineatematerialthat'snotoriginalbutisneededforthepaper.Beconcise--rememberthecriticalruleofthumb.
∙Content:
Themeatofthepaperincludesalgorithms,systemdescriptions,newlanguageconstructs,analyses,etc.Wheneverpossibleusea"top-down"description:
readersshouldbeabletoseewherethematerialisgoing,andtheyshouldbeabletoskipaheadandstillgettheidea.
PerformanceExperiments
WecouldhaveanentiretreatiseonthistopicaloneandIamsurelynottheexpert.Herearesomerandomthoughts:
∙Manyconferencesexpectexperiments.
∙It'seasytodo"hokey"ormeaninglessexperiments,andmanypapersdo.
∙It'seasytocraftexperimentstoshowyourworkinitsbestlight,andmostpapersdo.
∙Whatshouldperformanceexperimentsmeasure?
Possiblities:
oPurerunningtime
oSensitivitytoimportantparameters
oScalabilityinvariousaspects:
datasize,problemcomplexity,...
oOthers?
∙Whatshouldperformanceexperimentsshow?
Possibilities:
oAbsoluteperformance(i.e.,it'sacceptable/usable)
oRelativeperformancetonaiveapproaches
oRelativeperformancetopreviousapproaches
oRelativeperformanceamongdifferentproposedapproaches
oOthers?
TheConclusions
Ingeneralashortsummarizingparagraphwilldo,andundernocircumstancesshouldtheparagraphsimplyrepeatmaterialfromtheAbstractorIntroduction.Insomecasesit'spossibletonowmaketheoriginalclaimsmoreconcrete,e.g.,byreferringtoquantitativeperformanceresults.
FutureWork
Thismaterialisimportant--partofthevalueofapaperisshowinghowtheworksetsnewresearchdirections.Ilikebulletlistshere.(ActuallyIlikethemingeneral.)Acoupleofthingstokeepinmind:
∙Ifyou'reactivelyengagedinfollow-upwork,sayso.E.g.:
"Wearecurrentlyextendingthealgorithmto...blahblah,andpreliminaryresultsareencouraging."Thisstatementservestomarkyourterritory.
∙Conversely,beawarethatsomeresearcherslooktoFutureWorksectionsforresearchtopics.Myopinionisthatthere'snothingwrongwiththat--consideritacomplement.
TheAcknowledgements
Don'tforgetthemoryou'llhavepeoplewithhurtfeelings.Acknowledgeanyonewhocontributedinanyway:
throughdiscussions,feedbackondrafts,implementation,etc.Ifindoubtaboutwhethertoincludesomeone,includethem.
Citations
Spendtheefforttomakeallcitationscompleteandconsistent.DonotjustcopyrandominconsistentBibTex(orother)entriesfromthewebandcallitaday.Checkoveryourfinalbibliographycarefullyandmakesureeveryentrylooksright.
Appendices
Appendicesshouldcontaindetailedproofsandalgorithmsonly.Appendicescanbecrucialforoverlengthpapers,butarestillusefulotherwise.Thinkofappendicesasrandom-accesssubstantiationofunderlyinggorydetails.Asaruleofthumb:
∙Appendicesshouldnotcontainanymaterialnecessaryforunderstandingthecontributionsofthepaper.
∙Appendicesshouldcontainallmaterialthatmostreaderswouldnotbeinterestedin.
GrammarandSmall-ScalePresentationIssues
∙IngeneraleveryonewritingpapersisstronglyencouragedtoreadtheshortandveryusefulTheElementsofStylebyStrunkandWhite.Here'sarandomlistofpetpeeves.
∙Justlikeaprogram,all"variables"(terminologyandnotation)inthepapershouldbedefinedbeforebeingused,andshouldbedefinedonlyonce.(Exception:
Sometimesafteralonghiatusit'susefultoremindthereaderofadefinition.)Globaldefinitionsshouldbegroupedi
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 国外 应用文 学术论文 格式