语法分析器
l2期李明琴等:中文语义依存关系分析的统计模型l68lChinese(wi/si):杨/Ddl5博士/Ael3近年来/Caii十分/Kaol重视/Gb2l其/Aao4发明/Hco5成果/Dal4的/Kdol推广/Ieol使用/Hj28 Engiish:Theseyears,DoctorYangpaysaiotofattentiontothepopuiarizationandappiicationofhisinvention.
(a)Thesentencetaggedwithsemanticciassesi

Modifier(i)
Index
l
2
3
4
5
6
7
8
9
lo
ll杨/Yang博士/Doctor近年来/theseyears十分/aiot重视/payattentionto其/his发明/invention成果/production的/of推广/popuiarization使用/appiicationWordIndex2555-l88lo85loHeadWord(Hi)Word博士/Doctor重视/payattentionto重视/payattentionto重视/payattentionto-l成果/production成果/production推广/popuiarization成果/product重视/payattentionto推广/popuiarizationSR(i)SemanticReiation(Ri)限定/Restrictive经验者/Experiencer时间/Time程度/Degree核心成分/Kerneiword限定/Restrictive限定/Restrictive内容/Contentdependency“的”字依存‘/De’目标/Target连接/Coordination
(b)Thesentenceannotatedwithsemanticdependency
(c)Thesemanticdependencytreeofthesentence,headwordsareiinkedwithboidiines,andmodifierwordsareiinkedwitharrowiines
Figurel Anoverviewoftherepresentationusedbythemodei
Thetagsetofdependencyreiationscontains7o
[9]reiations,whichinciude59semanticreiations,9
syntacticreiationsand2speciaireiations.Mostofse-
manticreiationsareseiectedfromthesetofsemantic
[l3],forinstance,Agent,reiationsdefinedinHownet
Patient,Possessor,Possession,Experiencer,Content,
Time,Timelni,TimeFin,Location,Locationlni,Location-
Fin,Cause,Purpose,andsoon.Syntacticreiationsare
usedtoannotatethespeciaistructuresthatdonothave
‘De’exactsenseintermsofsemantics,forinstance,
dependency,Tense&voice,Preposition,andsoon.In
“Kerneladdition,therearetwospeciaireiations:
word”istoindicatetheheadwordofasentence,and
“Failure”istoindicatethewordthatcannotbeanno-
tatedwithdependencyreiationsbecausethesentenceis
notcompieted.
3 Statisticalmodel
3.1 Parsingprocedure
Howtoparseasentencewithdependencygram-
marwiiibedescribedbeforediscussingthestatisticai
modeiandaigorithm.Tobeginparsing,pairsofneigh-
borwordsthataredependentaremergedfirstiytoex-tendiargerconstituents,andthesenewconstituentsareiabeiedwiththeheadwordsandtheirmergingopera-tions(Forconvenience,theyarecaiiedasthehead-wordandtheiastoperationoftheconstituents,respec-tiveiy).Thenthenewconstituentsaremergedwithotherdependentneighborconstituentsorwords,andsoon,untiiaconstituentspanningthewhoiesentenceiscreated.Themergingoperationcanbewrittenintheformatof(SemanticReiation,ieft/right),where“Se-manticReiation”isthereiationbetweenthetwomer-gedwords,anditcanbeanyreiationdefinedinsec-tion2.2.“Left/right”indicatestheoriginofthehead-word,thatis,“Left/right”meanstheheadwordcomesfromtheieftorrightnode.Especiaiiy,NULLoperationisdefinedtodescribetheiastoperationforieavesofparsetree.Forexampie,themergingprocedurecanbedescribedasabinarytreeinFigure2.Inthebinarysemanticdependencytree,ieafnode“FaMing/inven-tion”isfirstiymergedwithieafnode“ChengGuo/pro-duction”tocreateanewconstituentiabeiedwith“ChengGuo/production”and(Restrictioe,right).,andsoon.Thenthenodeismergedwith“Oi/his”Sometimestwowordsaredirectiydependent
中文语义依存关系分析的统计模型_英文__文学_高等教育_教育专区 暂无评价0人阅读0次下载举报文档 中文语义依存关系分析的统计模型_英文__文学_高等教育_教育...
基于配价结构和语义依存关系的句法分析统计模型_天文/地理_自然科学_专业资料 暂无评价
中南大学 信息科学与工程学院,湖南 长沙,410083) 摘要:利用语义、语法等语言知识,建立一种基于依存关系的句法分析统计模型,并利用改进的句法分析模型 进行句法分析...
语共现统计的方法是一种最基本 的方 的基 本 ...斯 坦福 大学 自然语 言 句子语义依存关系分析可以...(6):531—534.(in Chinese) 模型进行 了 比较 ...
,主要存在两种方法 : ①基于向量空间模型的方法 。...的句法分析 , 找出依存关系 , 并在 依存分析结果...下文中的统计特性 ,而没有考虑词本身的语义信息 。...
汉语语义依存分析_计算机软件及应用_IT/计算机_专业资料。汉语句法分析方面的文章...汉语语义分析模型研究述... 7页 1下载券 基于树库的汉语依存句法... 5页 ...
关键词: 计算机应用; 中文信息处理; 语料库; 语义依存关系;5 知网6; 动态...类分得越细, 统计数据的稀疏问题越严重, 会对分析模型的鲁棒性造成不良 影响。...
原信息为特征,基于该特征训练得到支持 向量机(SVM)依存分析模型,从而有效识别出了术语内部的依存关系;在依存分析基础上利用统计和 规则相结合的方法进行术语语义分析...
汉语语义分析模型研究述评_文学_高等教育_教育专区。汉语语义分析模型研究述评维普...中文语义依存关系分析的... 11页 免费 基于统计学习模型的句法... 12页 免费...
http://research.microsoft.com/~jfgao/paper/ 附中文参考文献: [10] [11] [13] [14] 李明琴,李涓子,王作英,陆大纟 .中文语义依存关系分析的统计模型....
热门文档