题名:
英语语料库与自动语法分析   / 方称宇 ,
ISBN:
978-7-100-05659-5 价格： CNY38.00
语种:
中
载体形态:
225页 16开
出版发行:
出版地： 出版社： 商务印书馆 出版日期： 2007.11
内容提要:
语料库语言学和计算语言学为促进自然语言处理技术快速发展的两门基础学科。《英语语料库与自动语法分析》系这两个领域的一本专著,它以国际英语语料库为背景,着重探讨大型语料库的语法分析,尤其是英语口语材料给计算机自动处理带来的一系列难题。书中涉及基于概率的自动词类识别和基于实例的自动句法分析这两大技术,并有专门章节来探讨句法分析的评测问题,对AUTASYS和The Survey Parser这两个软件系统的实际表现进行了深入的量化评测。此外,本书还探讨了介词短语的自动分析,特别是这类短语的句法功能的自动判定,并对自动语法分析在语音合成及语音识别中的应用做了相应的说明。 
主题词:
英语  
中图分类法:
H31 版次： 4
主要责任者:
方称宇
版次:
1
内容附注:
Preface前言List of FiguresList of TablesAbstract1. Introduction1.1. What is Parsing?1.2. The Introspective View1.3. The Retrospective View1.4. Data-Oriented Parsing1.5. General Problems1.6. The Proposed Research1.6.1. Background to the Proposed Research1.6.2. The Basic Approach of the Proposed Research1.6.3. The Strengths and Novelties of the Proposed Approach1.6.3.1. Automated Grammar Generation1.6.3.2. De-Lexicalised Terminal Nodes1.6.3.3. Global Parse with Subcategorisation Features1.6.3.4. High-Quality Partial Parse1.6.3.5. Intrinsic Ability to Learn1.7. The Organisation of the Book2. The Automatic Analysis of English Word Classes2.1. An Overview of Word Class Tagging2.2. Major Word Class Tagging Schemes2.2.1. The Lancaster-Oslo/Bergen Tagging Scheme2.2.1.1. The Lancaster-Oslo-Bergen Corpus2.2.1.2. The Lancaster-Oslo-Bergen Tag Set2.2.1.3. Summary2.2.2. The International Corpus of English Tagging Scheme2.2.2.1. The International Corpus of English2.2.2.2. The International Corpus of English Tag Set2.2.3. A Comparison of LOB and ICE2.3. Word Class Tagging Methodologies2.3.1. The Rule-Based Approach2.3.2. The Probabilistic Approach2.4. AUTASYS: A Hybrid Tagging System2.4.1. A Probabilistic Approach Using the LOB Tag Set2.4.1.1. The Tag Assignment Module2.4.1.1.1. Tokenisation2.4.1.1.2. The treatment of"."2.4.1.1.3. The treatment of"'"2.4.1.1.4. Sentence boundary markers2.4.1.2. Orthographic Analysis2.4.1.3. Lexicon Lookup2.4.1.3.1. The lexicon2.4.1.3.2. The coverage of the lexicon2.4.1.4. Morphological Analysis2.4.2. The Idiom Identification Module2.4.3. The Probabilistic Tag Selection Module2.4.3.1. The Bigram Probabilistic Matrix2.4.3.2. Implementing Probabilistic Tag Selection2.4.4. The Rule-Based Refinement Module2.4.5. Empirical Evaluation2.4.6. Permissive AUTASYS-LOB Disagreements2.4.6.1. NNP-NPT2.4.6.2. JJ-JJB2.4.6.3. NNP-NPL2.4.6.4. RB-NN2.4.7. Summary2.5. A Rule-Based Approach towards LOB to ICE Translation2.5.1. Solutions for Verbs2.5.1.1. Auxiliary vs. Lexical2.5.1.2. Monotransitive vs. Complex Transitive2.5.1.3. Finite vs. Nonfinite2.5.2. Closed Sets2.5.3. Initial Results2.5.4. Problems2.5.5. Summary3. The Automatic Induction of a Formal Grammar3.1. Introduction3.2. The ICE Parsing Scheme3.3. Grammar Generation3.3.1. Phrase Structure Rules3.3.2. Phrase Structure Cluster Rules3.4. Evaluating the Coverage3.4.1. The Construction of the Training and Test Sets3.4.2. The Number of Extracted Rules3.4.3. The Coverage of Adjective Phrase Rules3.4.4. The Coverage of Adverb Phrase Rules3.4.5. The Coverage of Noun Phrase Rules3.4.6. The Coverage of Verb Phrase Rules3.4.7. The Coverage of Prepositional Phrase Rules3.4.8. The Coverage of Phrase Structure Cluster Rules3.4.9. Discussion4. Robust Practical Analogy-Based Parsing4.1. Introduction4.1.1. Analogy-Based Parsing4.2. An Overview of the Survey Parser4.3. The Construction of the Syntactic Knowledge Base4.3.1. Phrase Structure Rules4.3.2. Phrase Structure Cluster Rules4.3.2.1. Feature Constraints4.3.2.2. Nonobligatory Elements4.3.2.3. A Definition of Analogy4.4. Parsing with Phrase Structure and Phrase Structure Cluster Rules4.4.1. The Analysis of Word Classes4.4.2. The Analysis of Phrases4.4.3. The Analysis of Clauses4.4.3.1. Partial Analysis4.5. Some Initial Evaluation4.5.1. Evaluating the Coverage of Phrase Structure Rules4.5.2. Evaluating the Coverage of Phrase Structure Cluster Rules4.5.3. Evaluating the Labelling Precision4.5.4. Evaluating the Accuracy of Analysis4.5.5. Evaluating the Processing Speed4.6. Concluding Remarks5. Extensive Evaluations of the Survey Parser5.1. Introduction5.2. Commonly Used Metrics5.2.1. Labelled Match5.2.2. Bracketed Match5.2.3. Crossing-Brackets Rate5.2.4. Summary5.3. Evaluations with the NIST Scheme5.3.1. Labelling Accuracy5.3.1.1. Methodology5.3.1.2. Evaluating the Scoring Program5.3.1.3. Evaluating the Labelling Accuracy of Parser Produced Trees5.3.2. Bracketing Accuracy5.3.2.1. A Linear Representation of the Hierarchical Structure5.3.2.2. Evaluating the Scoring Program5.3.2.3. Evaluating the Bracketing Accuracy of Parser Produced Trees5.3.3. Labelling and Bracketing Accuracy Scores Combined5.3.3.1. A Description5.3.3.2. Empirical Scores by the Survey Parser5.3.3.3. A Comparison with other Reports5.3.3.3.1. A Comparison with Keller (2003)5.3.3.3.2. A Comparison with Plaehn (2004)5.3.3.3.3. A Comparison with Henderson (2004)5.3.3.3.4. Summary5.4. Summary6. The Resolution of Prepositional Phrases6.1. Introduction6.2. PPs in Contemporary British English6.2.1. Prepositions6.2.2. Prepositional Complement6.2.3. The Syntactic Functions of Prepositional Phrases6.3. Data and Resources Used in the Experiment6.4. Scope of Experiment6.5. Test Data6.6. Lexical Database6.7. Rules and their Coverage6.7.1. Prepositional Phrases as Adjective Phrase Postmodifiers6.7.1.1. Test Data6.7.1.2. Attachment Rules6.7.1.3. Morphological Analysis6.7.1.4. Coverage6.7.2. Prepositional Phrases as Noun Phrase Postmodifiers6.7.2.1. Test Data6.7.2.2. Attachment Rules6.7.2.3. Morphological Analysis6.7.2.4. Coverage of the Rules6.7.3. Prepositional Phrases as Adverbials6.8. Overall Success Rate and Discussion7. Conclusions and Further Work7.1. Conclusions7.2. Applications of AUTASYS and the Survey Parser7.2.1. AUTASYS7.2.2. The Survey Parser7.2.2.1. SpeechMaker7.2.2.2. Correlation between Tone Units and Syntax7.3. Future Work7.3.1. Implementation of Automated Prepositional Phrase Attachment7.3.2. Automated Clause Boundary Detection and AttachmentReferencesAppendix A: A List of LOB TagsAppendix B: A List of ICE TagsAppendix C: A List of AUTASYS IdiomsAppendix D: A List of ICE Parsing SymbolsAppendix E: A List of ICE Prepositions in Descending Frequency OrderAppendix F: A Distributional Profile of ICE-GB PrepositionsIndex