北野武(north

《北野武(north》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、UsingtheWebforAutomatedTranslationExtractioninCross-LanguageInformationRetrievalAdvisor:Dr.HsuPresenter:Zih-HuiLinAuthor:YingZhangandPhilVines1MotivationObjectivePreviousworkMethodologyExperimentsandresultsConclusionsOutline2MotivationOneofthemajorremainingreasonsthatCLIRdoesnotperformaswellasmonoli

2、ngualretrievalisthepresenceofoutofvocabulary(OOV)terms.itwillnotberecognized,andsegmentedintoeithersmallersequencesofcharactersorindividualcharacters北野武→(northlimitmilitary)Previousworkhaseitherreliedonmanualinterventionorhasonlybeenpartiallysuccessfulinsolvingthisproblem.3ObjectiveWeproposeasegment

3、ationfreemethodwhichcanbeappliedtobothChinese-EnglishandEnglish-ChineseCLIR,correctlyextractingtranslationsofOOVtermsfromtheWebautomatically,andthusisasignificantimprovementonearlierwork4EnglishtranslationextractioninChinese-EnglishCLIRChineseOOVtermdetection北野武(northlimitmilitary)→PvaluegivenbytheH

4、MMwillbeverylowifPvalue

5、Ctargets

6、=max(

7、Cij

8、

9、).2.f(et,Ct)=max(f(ei,Ctargets)).3.Add(Ct,et)intothetranslationdictionary.1.f(etargets)=max(f(ei)).2.f(et’,Ct’)=max(f(etargets,Cij)).3.ifCt’≠Ctandet’≠et,add(Ct’,et’)intothetranslationdictionary.北野武（KitanoTakeshi）c4c5c6e1導演北野武（KitanoTakeshi）c2c3c4c5c6e15ChinesetranslationextractioninEnglish-ChineseC

10、LIRExtractionofwebtextuseGoogletofetchthetop100ChinesedocumentswiththeEnglishOOVtermeoovasthequery.Collectionofco-occurrencestatisticsaccumulatethefrequencyfoov.consideringallsubstringsinSleftandSright,andcollectingthefrequencyfnandthelength

11、sn

12、ofeachChinesesubstring.Translationselectionexcludeanysu

13、bstringthatalreadyinthetranslationdictionarydoesn’toccurinthedocumentcollection與區域貿易………兩岸經貿關係（CanadaandCrossStraitsEconomicRelations）三組發表九………英茂、加Sleft(包含20個字)Sright(包含20個字)eoovLongestlengthHighestfreq