On Natural Language Processing and Plan Recognition

On Natural Language Processing and Plan Recognition

ID:38147585

大小:109.58 KB

页数:6页

时间:2019-05-27

On Natural Language Processing and Plan Recognition_第1页
On Natural Language Processing and Plan Recognition_第2页
On Natural Language Processing and Plan Recognition_第3页
On Natural Language Processing and Plan Recognition_第4页
On Natural Language Processing and Plan Recognition_第5页
资源描述:

《On Natural Language Processing and Plan Recognition》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、OnNaturalLanguageProcessingandPlanRecognitionChristopherW.GeibandMarkSteedmanSchoolofInformaticsUniversityofEdinburghEdinburghEH89LW,Scotlandcgeib@inf.ed.ac.ukAbstracttheseareindividualwordsorutterances.Inbothcases,theobservationsareusedtocreateahigherlevelstructure.InTheresearchareaso

2、fplanrecognitionandnatu-NLPthesehigherlevelstructuresmaybeparsetrees[Collins,rallanguageparsingsharemanycommonfeatures1997]orlogicalforms[Bosetal.,2004].InPRtheyareandevenalgorithms.However,thedialogbetweenusuallyahierarchicalplanstructure[KautzandAllen,1986;thesetwodisciplineshasnotbe

3、eneffective.Specif-Kaminkaetal.,2001;GeibandGoldman,2003]oratleastaically,significantrecentresultsinparsingmildlyhighlevelrootgoal[Horvitzetal.,1998].Ineithercase,bothcontextsensitivegrammarshavenotbeenlever-NLPandPRconstructahigherlevelknowledgestructurethatagedinthestateoftheartplanrec

4、ognitionsys-relatesthemeaningsofeachoftheindividualobservationstotems.Thispaperwilloutlinetherelationsbetweenameaningforthecollectionofobservationsasawhole.naturallanguageprocessing(NLP)andplanrecog-Forthepurposesofthisdiscussionitwillaidustoabstractnition(PR),arguethateachofthemcaneffe

5、ctivelyawayfromthespecificdetailsofthehigherlevelstructurethatinformtheother,andthenfocusonkeyrecentre-isbuiltbythisprocess.TosimplifythisdiscussionwewillsearchresultsinNLPandarguefortheirapplica-talkaboutthesesystemsasiftheywerecreatinganhierarchi-bilitytoPR.caldatastructurethatcapture

6、sthemeaningofthecollectionofobservations.WewillusethePRterminologyandcallthis1IntroductionstructureanexplanationandfollowingtheNLPterminologycalltheprocessofproducingasingleexplanationparsing.Withoutperformingacarefulliteraturesearchonecouldeas-Inordertoparseasetofobservationsintoanexp

7、lanationilyimaginethatthefieldsofPlanRecognition(PR)andNat-bothPRandNLPmustspecifythepatternsofobservationsuralLanguageProcessing(NLP)aretwoseparatefieldsthattheyarewillingtoacceptortherulesthatgovernhowtheob-havelittleincommon.Therearefewpapersineitherdisci-servationscanbecombined.InP

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。