基于图的统计机器翻译词汇化调序模型(英文)-豆柴文库

您所在位置：网站首页 / 基于图的统计机器翻译词汇化调序模型(英文).docx / 文档详情

在线预览结束，喜欢就下载吧，查找使用更方便

5 金币

下载文档

/ 2

下载提示

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

文本预览

基于图的统计机器翻译词汇化调序模型(英文)
Introduction:
Asmachinetranslationhasbecomemoreadvanced,thefocushasshiftedfromsimplytranslatingthewordstoworkingonimprovingthequalityofthetranslation.Oneofthemainissueswithmachinetranslationisthedifficultyinaccuratelytranslatingthewordorderfromthesourcelanguagetothetargetlanguage.Toaddressthisproblem,researchershavebeenexploringtheuseofgraph-basedstatisticalmachinetranslationmodelsthatincorporatelexicalizedreordering.
Thispaperwilldiscussthedevelopmentofagraph-basedstatisticalmachinetranslationmodelthatincorporateslexicalizedreorderingtoimprovetheaccuracyofwordordertranslationfromthesourcelanguagetothetargetlanguage.
Background:
Machinetranslationistheprocessofusingcomputerstotranslatetextfromonelanguagetoanother.Althoughmachinetranslationhasbeenaroundfordecades,itisonlyinrecentyearsthatithasbecomemoreadvancedandwidelyused.Therearetwomainapproachestomachinetranslation:rule-basedandstatisticalmachinetranslation.Whilerule-basedmachinetranslationreliesonasetofpredefinedrulesandgrammar,statisticalmachinetranslationuseslargeamountsofdatatolearnhowtotranslate.
Oneofthemainchallengesofstatisticalmachinetranslationisaccuratelytranslatingtheorderofwordsfromthesourcelanguagetothetargetlanguage.Differentlanguageshavedifferentsentencestructures,anditcanbedifficultforacomputertoaccuratelytranslatetheorderofwordswhilestillmaintainingthemeaningofthesentence.Therefore,researchershavebeenexploringwaystoincorporatelexicalizedreorderingtoimprovemachinetranslation.
Methods:
Inthisstudy,wedevelopedagraph-basedstatisticalmachinetranslationmodelthatincorporateslexicalizedreordering.Themodelusesadirectedgraphtorepresentthesourcelanguagesentence,witheachnoderepresentingawordinthesentence.Thegraphisconstructedbyconsideringboththelexicalandstructuralinformationinthesentence.
Wealsoincorporatedalexicalizedreorderingmodelthatisabletolearnthepositionsofwordsinasentencefromthetrainingdataset.Thismodelisusedtoreordertheverticesinthegraphtorepresentthemostlikelytargetlanguages