Abstract Shallow NLP techniques for Internet Search

Abstract Shallow NLP techniques for Internet Search

ID:37656869

大小:412.38 KB

页数:10页

时间:2019-05-27

Abstract Shallow NLP techniques for Internet Search_第1页
Abstract Shallow NLP techniques for Internet Search_第2页
Abstract Shallow NLP techniques for Internet Search_第3页
Abstract Shallow NLP techniques for Internet Search_第4页
Abstract Shallow NLP techniques for Internet Search_第5页
资源描述:

《Abstract Shallow NLP techniques for Internet Search》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、ShallowNLPtechniquesforInternetSearchAlexPenevRaymondWongNationalICTAustraliaandSchoolofComputerScienceandEngineering,UniversityofNewSouthWales,Sydney,NSW2052,Australia{alexpenev,wong}@cse.unsw.edu.auAbstractoraustriancomposersborninthe1600s”and“na-tiveanimalsinaus

2、tralia,butnotmarsupials”showInformationRetrieval(IR)isamajorcomponentinthatthereisroomforimprovementintoday’sen-manyofourdailyactivities,withperhapsitsmostgines.Moreelaboratequeriessuchas“700mlJohnnieprominentrolemanifestedinsearchengines.Today’sWalkerRedLabel,inSy

3、dneyforunder$30”cannotmostadvancedenginesusethekeyword-based(“bagbeansweredatall,evenwhenappropriatedocumentsofwords”)paradigm,whichconcedessomeinher-areindexedbytheengine.entdisadvantages.WebelievethatnaturallanguageThroughpracticeandtribulation,userslearnto(NL)is

4、amoreuser-oriented,context-preservativementallyculltheirqueriesintoasetofonlythefewandintuitivemechanismforwebsearch.“mostimportant”components,beforesendingare-Inthispaper,weexploreshallowNLPtechniquesquesttoanengine.Thisstructuralrearrangement,tosupportarangeofNLq

5、ueriesoveranexistingcoupledwithsearchengines’treatmentofkeywordskeyword-basedengine.WepresentJASE,awebap-asindividuals,couldbeimpedingtheiraccuracy.plicationenvelopingtheGooglesearchengine,whichFurthermore,tobecomeproficientatusingacer-performswebsearchesbydecomposi

6、nginputNLtainengine,usersmustlearnitsspecialoperators.queriesandgeneratingnewqueriesthataremoreThesedifferbetweeneachengineandrenderthesuitableforthesearchengine.Byusingsomeofsearchmechanismtobeunnatural,duetothein-Google’ssyntacticoperatorsandfilters,itcreatestroduc

7、tionofforeignmodifiersintothequery.Our“clever”queriestoimproveprecision.evaluationsurveyindicatesthataverageGoogleusersApreliminaryevaluationwasconductedtotestarelargelyunawareofitsoperatorsandfilters,andJASE’saccuracy,andresultshavebeenencouraging.rarelyusetheminpra

8、ctice.WeconcludethattheNLmodelhaspotentialtonotTherehasbeenfewsignificantadvancesinInter-onlyrivalthekeyword-basedparadigm,butsubstan-netsearchfor

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。