學(xué)者們對(duì)于回指研究的熱情已經(jīng)歷經(jīng)半個(gè)世紀(jì)而不衰,研究方法也涉及各種語(yǔ)言學(xué)理論的方方面面。指代消解研究起源于自然語(yǔ)言處理,其主要目的是使用計(jì)算機(jī)來(lái)為篇章中的回指語(yǔ)找到正確的指代對(duì)象。因其從計(jì)算語(yǔ)言學(xué)的角度著手,著重研究回指的生成和理解乃至篇章的理解,近年來(lái)成為回指研究的新方向,引起了不少學(xué)者的關(guān)注,但這方面的對(duì)比研究幾乎還為空白,有待填補(bǔ)。指代消解研究所采用的理論框架是多種多樣的,向心理論是其中的一種。在向心理論的基礎(chǔ)上進(jìn)行指代消解開(kāi)創(chuàng)了指代消解的一個(gè)新局面,因?yàn)橄蛐睦碚撉〉胶锰幍匕颜Z(yǔ)言學(xué)理論框架和計(jì)算機(jī)系統(tǒng)要求的可操作性結(jié)合在一起。向心理論描述了話(huà)語(yǔ)的注意焦點(diǎn)、指稱(chēng)方式的選擇和語(yǔ)篇連貫性之間的關(guān)系,利用“中心”概念對(duì)語(yǔ)篇的連貫性和顯著性進(jìn)行解釋,并為讀者（聽(tīng)者）識(shí)別代詞或名詞詞組的指稱(chēng)對(duì)象提供了三條制約條件和兩條規(guī)則。其中制約條件三和規(guī)則一為語(yǔ)篇中的回指語(yǔ)提供了一個(gè)理想的消解模式。向心理論是二十世紀(jì)八十年代初由計(jì)算語(yǔ)言學(xué)領(lǐng)域的學(xué)者提出的,之后被廣泛的應(yīng)用于各種語(yǔ)言的指代消解和篇章處理中。本文選擇該理論作為研究基礎(chǔ)一是因?yàn)樗谥复夥矫骟w現(xiàn)出的有效性,二是因?yàn)橐韵蛐睦碚摓榛A(chǔ)的對(duì)比框架... 

【文章來(lái)源】：上海外國(guó)語(yǔ)大學(xué)上海市 211工程院校 教育部直屬院校

【文章頁(yè)數(shù)】：200 頁(yè)

【學(xué)位級(jí)別】：博士

【文章目錄】：
Acknowledgement
Abstract
摘要
Table of Contents
List of Tables
List of Figures
Chapter 1 Introduction
    1.1 Motivations
    1.2 Scope of the Study
    1.3 Methodology , Hypotheses and Objectives of the Study
    1.4 Organization of the Dissertation
Chapter 2 Centering Theory and Its Framework
    2.1 Introduction
    2.2 Origin and Development
        2.2.1 Original Intuitions and Intentions
            2.2.1.1 Focus
                2.2.1.1.1 Grosz （1977）
                2.2.1.1.2 Sidner （1979, 1981, 1983）
                2.2.1.1.3 Grosz and Sidner （1985, 1986）
            2.2.1.2 Center
        2.2.2 Formulation
    2.3 Framework
        2.3.1 Terminology and Definition
            2.3.1.1 Discourse Segment （D） and Utterance （U）
            2.3.1.2 Centers, Forward-Looking Centers and Backward-Looking Centers
            2.3.1.3 Ranking and Preferred Centers
            2.3.1.4 Realization
        2.3.2 Constraints and Rules
        2.3.3 Basic Claims and Issues in Centering
            2.3.3.1 Basic Claims
            2.3.3.2 Local Coherence and Local Salience
                2.3.3.2.1 Local Coherence
                2.3.3.2.2 Local Salience
    2.4 Centering as a Cross-linguistic and Language-specific Theory
    2.5 Centering as a Parametric Theory
        2.5.1 The Parameters of Centering
            2.5.1.1 Utterance （Ui） and Previous Utterance （Ui-1）
            2.5.1.2 Ranking
            2.5.1.3 Rule 1 Pronouns
            2.5.1.4 Realization
            2.5.1.5 Discourse Segmentation
Chapter 3 Centering and Anaphora Resolution
    3.1 Introduction
    3.2 Influential Algorithms of Anaphora Resolution in Literature
        3.2.1 Tree Search Algorithm / Hobbs’Algorithm
            3.2.1.1 The Algorithm
            3.2.1.2 Comments
        3.2.2 RAP （Resolution of Anaphora Procedure）
            3.2.2.1 Filters and Components in RAP
            3.2.2.2 Algorithm in RAP
            3.2.2.3 Comments
        3.2.3 Robust, Knowledge-poor Approach
            3.2.3.1 Working Mechanism
            3.2.3.2 Comments
    3.3 Strategies and Theories
        3.3.1 Computing Strategies
        3.3.2 Theoretical Framework
    3.4 Centering Algorithms
        3.4.1 BFP
            3.4.1.1 Algorithm
            3.4.1.2 Comments
        3.4.2 LRC （Left-Right Centering Algorithm）
            3.4.2.1 Algorithm
            3.4.2.2 Comments
        3.4.3 Optimization Theory
            3.4.3.1 Algorithm
            3.4.3.2 Comments
    3.5 Algorithm Design in China
        3.5.1 Centering-Based Algorithms
            3.5.1.1 Yeh and Chen （2001, 2003）
                3.5.1.1.1 Algorithms
                3.5.1.1.2 Comments
            3.5.1.2 Wang （2004）
                3.5.1.2.1 Algorithms
                3.5.1.2.2 Comments
            3.5.1.3 Xu, Duan and Fu （2006, 2008, 2009）
                3.5.1.3.1 Algorithms
                3.5.1.3.2 Comments
Chapter 4 Methodology of the Present Study
    4.1 Introduction
    4.2 Poesio’s Empirical Study of Parametric Centering
        4.2.1 Parameter Setting
        4.2.2 Annotation of Corpus
            4.2.2.1 Utterance
            4.2.2.2 NPs
            4.2.2.3 Anaphoric Information
            4.2.2.4 Segmentation
        4.2.3 Results and Enlightenment
            4.2.3.1 Utterance
            4.2.3.2 Realization
            4.2.3.3 Segmentation
            4.2.3.4 Ranking
            4.2.3.5 Rule-1 Pronoun
            4.2.3.6 Difference Between Domains
    4.3 Contrastive Evaluation of Centering Parameters
        4.3.1. Segmentation
        4.3.2 Rule-1 Pronouns
        4.3.3 Utterance
        4.3.4 Realization
        4.3.5 Ranking
            4.3.5.1 Theoretical Basis
            4.3.5.2 Topic in Chinese
            4.3.5.3.R anking Scale by Grammatical Roles in our Research
            4.3.5.4 Other Means of Ranking
Chapter 5 Corpus and Annotation
    5.1 Introduction
    5.2 Corpus
    5.3 Annotation
        5.3.1 Basis of Annotation: Penn Treebank
        5.3.2 Adaptation to Our Study
            5.3.2.1 Noun Phrase
                5.3.2.1.1 NP Types
                5.3.2.1.2 Grammatical and Morphological Information
            5.3.2.2 Grammatical Roles
            5.3.2.3 Hierarchical Structures
            5.3.2.4 Reference Types
            5.3.2.5 Positions of Antecedents and database
        5.3.3 Open Issues in Annotation
            5.3.3.1 Coordinated NP
            5.3.3.2 “It”as Surface Subject
            5.3.3.3 Special Sentence Patterns
Chapter 6 Algorithms in our Research
    6.1 Introduction
    6.2 Guiding Principles
        6.2.1 Eliminating Principles
            6.2.1.1 Morphological Filter
            6.2.1.2 Syntactic Filter
            6.2.1.3 Inaccessibility Inheritance Principle
        6.2.2 Preference Principles
    6.3 Algorithms in our Study
        6.3.1 Inter-algorithm Operating Procedure
        6.3.2 Inner-algorithm Operating Procedure
        6.3.3 Algorithms and their Structure in our research
            6.3.3.1 Alg.1 Lin
            6.3.3.2 Alg.2 Grm
            6.3.3.3 Alg.3 Para
            6.3.3.4 Alg.4 Cb
            6.3.3.5 Alg.5 Para +Cb
            6.3.3.6 Alg.6 Sub
Chapter 7 Results Analysis and Discussion
    7.1 Introduction
    7.2 Preliminaries
        7.2.1 Statistical Overview of the Corpus
        7.2.2 Terms: Recall, Success and Precision Rates
        7.2.3 Other Abbreviations
    7.3 Results in Terms of Utterance
        7.3.1 Chinese
        7.3.2 English
    7.4 Results in Terms of Rule-1 Pronouns
        7.4.1 Chinese
        7.4.2 English
    7.5 Results in Terms of Ranking
        7.5.1 Linear Order
        7.5.2 Grammatical Function
        7.5.3 Other Factors Coupled with Grammatical Function
            7.5.3.1 Chinese
            7.5.3.2 English
        7.5.4 Main/Surbordinate Hierarchy
            7.5.4.1 Chinese
            7.5.4.2 English
    7.6 Results in Terms of Zero Topic （Subject）
Chapter 8 Conclusion
    8.1 Summary of Major Findings
    8.2 Theoretical and Practical Implications of the Study
    8.3 Limitations of the Study and Suggestions for Future Research
References
Appendix Ⅰ Sample of Annotated Chinese Corpus
Appendix Ⅱ Sample of Annotated English Corpus
Appendix Ⅲ Interface for Database Information Input
Appendix Ⅳ Resolution Results of Chinese Corpus
Appendix Ⅴ Resolution Results of English Corpus


【參考文獻(xiàn)】：
期刊論文
[1]零形式和零成分的確立條件[J]. 袁毓林.  當(dāng)代語(yǔ)言學(xué). 2010(03)
[2]前瞻中心的排序?qū)χ复獾挠绊憽豁?xiàng)向心理論參數(shù)化實(shí)證研究[J]. 段嫚娟,許余龍,付相君.  外國(guó)語(yǔ)(上海外國(guó)語(yǔ)大學(xué)學(xué)報(bào)). 2009(03)
[3]向心理論的參數(shù)化研究[J]. 許余龍.  當(dāng)代語(yǔ)言學(xué). 2008(03)
[4]“語(yǔ)句”與“代詞”的設(shè)定對(duì)指代消解的影響——一項(xiàng)向心理論參數(shù)化實(shí)證研究[J]. 許余龍,段嫚娟,付相君.  現(xiàn)代外語(yǔ). 2008(02)
[5]中心理論和回指解析計(jì)算法[J]. 劉禮進(jìn).  外語(yǔ)學(xué)刊. 2005(06)
[6]漢語(yǔ)零形回指解析——基于向心理論的研究[J]. 王德亮.  現(xiàn)代外語(yǔ). 2004(04)
[7]語(yǔ)篇向心理論述評(píng)[J]. 苗興偉.  當(dāng)代語(yǔ)言學(xué). 2003(02)
[8]語(yǔ)篇回指的認(rèn)知語(yǔ)言學(xué)研究與驗(yàn)證[J]. 許余龍.  外國(guó)語(yǔ)(上海外國(guó)語(yǔ)大學(xué)學(xué)報(bào)). 2003(02)
[9]語(yǔ)篇回指的認(rèn)知語(yǔ)言學(xué)探索[J]. 許余龍.  外國(guó)語(yǔ)(上海外國(guó)語(yǔ)大學(xué)學(xué)報(bào)). 2002(01)
[10]Referring Expressions: A Unified Approach[J]. K.M. Jaszczolt.  外國(guó)語(yǔ)(上海外國(guó)語(yǔ)大學(xué)學(xué)報(bào)). 2001(02)



本文編號(hào)：3051469