基于語料庫的維吾爾語詞干提取和詞性標(biāo)注.pdf

1、新疆大學(xué)碩士學(xué)位論文基于語料庫的維吾爾語詞干提取和詞性標(biāo)注姓名：陳鵬申請學(xué)位級別：碩士專業(yè)：計算機應(yīng)用技術(shù)指導(dǎo)教師：古麗拉阿東別克20060611劉f柑凡學(xué)壩l。學(xué)位L侖衛(wèi)UyghurStemSegmentationandPOSTaggingbasedOiqCorporaAbstractConstructinghigh—qualitytamedcorporaisafundamentalpartinthefieldofUyghurnatu

2、rallanguageprocessingAtpresent，morecorporaofhigherqualityarerequiredinthefieldsofmachinetranslation(MT)，informationretrieval(IR)，webtextmining，etcAutomaticStemSegmentationandPartOfSpeech(POS)taggingarefundamentaltothecon

3、structionoftaggedCOrporaThisthesisintendstosolvetheproblemoftheStemSegmentationbycombiningtheBidirectionalMatchingalgorithmandOmni—wordSegmentationalgorithmComparedwiththeMaximumMatchingalgorithm，thismethodcanimprovethep

4、recisionofthestemsegmentationInthisthesis，theimprovedbinary—seek—by—characterdictionaryquerymechanismisemployedintheapplicationofUyghurstemsegmentationanditcanimprovetheefficiencyFurthermore，POStaggingmethodsareexplored，

5、andthemeritsanddemeritsofbothrule‘basedandstatistic—basedmethodsareanalyzedTheUyghurPOStaggingisstudiedbyapplyingtheprobabilisticmethodandtheunigramHiddenMaNovModel(HMM)isadoptedTheRelativeFrequencyTraining(RFr)methodisu

6、sedtoestimatethemodelparametersAndtheproblemofthedatasparsenessissolvedthroughthebackingoffdatasmoothingalgorithmAtlast，thepartofspeechistaggedinsentencesbyutilizingViterbialgorithmTheunigramHMMbasedontheprobabilisticmet