Machine translation : a knowledge-based approach

著者

書誌事項

Machine translation : a knowledge-based approach

Sergei Nirenburg ... [et al.]

Morgan Kaufmann Publishers, c1992

大学図書館所蔵 件 / 全19件

この図書・雑誌をさがす

注記

Bibliography: p. 225-248

Includes index

内容説明・目次

内容説明

This is the first book devoted exclusively to knowledge-based machine translation. While most approaches to the machine translation for natural languages seek ways to translate source language texts into target language texts without full understanding of the text, knowledge-based machine translation is based on extracting and representing the meaning of the source text. It is scientifically the most challenging approach to the task of machine translation, and significant progress has been achieved within it in recent years. The authors introduce the general paradigm of knowledge-based MT, survey major recent developments, compare it with other approaches and present a paradigmatic view of its component processes-natural language analysis, natural language generation, text meaning representation, ontological modeling, etc. Special chapters are devoted to machine-aided translation, speech translation, and challenges and solutions for knowledge representation. Based on these analyses, as well as on a review of general trends in MT, the authors discuss interesting directions for future research and development. This book will be of interest to researchers and advanced students in machine translation, natural language processing, computational linguistics, artificial intelligence, and even to some philosophers of language and theoretical linguists.

目次

Preface 1 MT in a Nutshell 1.1 Introduction 1.2 A Concise History of MT Research 1.3 The Mandate of Machine Translation, as Viewed by Users 1.3.1 What Do We Translate? 1.3.2 Tradeoffs in MT 1.3.2.1 Modifying the Acceptability Threshold 1.3.2.2 Partial Automation in MT 1.3.2.3 Restricting the Ambiguity of SL Text 1.3.3 Bilingual or Multilingual Machine Translation? 1.3.4 Text and Speech Translation 1.3.5 Evaluating the Quality of Translation 1.3.6 Quality and Utility of MT Systems 1.3.7 A More Comprehensive Set of MT Performance Metrics 1.4 Problems in MT, as Viewed by System Developers 1.4.1 Selected Problems in Text Understanding 1.4.2 Does an MT Program Really Need to Understand the Source Text Fully 1.5 Knowledge-Based Machine Translation 1.5.1 Transfer or Interlingua? 1.5.2 Nature and Size of Knowledge Bases 1.5.3 Human-Computer Interaction 1.5.4 MT as an Experimental Testbed for Computational Linguistics 1.6 Reader's Guide 2 Treatment of Meaning in MT Systems 2.1 Transfer versus Interlingua 2.2 On the Possibility of Translation 2.3 Understanding and Translation 2.4 Meaning Across Languages 2.5 Reasibility of General Meaning Representation 2.6 How Formal Must Meaning Representation Be? 2.7 Extractability of Meaning 2.8 Translation and Paraphrasing 2.9 Conclusion 3 The Concept of Interlingua 3.1 Requirements 3.2 Ontology 3.3 Representation of Meaning 3.3.1 Textual Meaning 3.3.2 The Goal and Plan Component 3.3.3 Speech Situation 3.3.4 Relations 3.3.4.1 Domain Relations 3.3.4.2 Text Relations 3.3.4.3 Intention-Domain Relations 3.3.5 Discussion 3.4 An Extended Example 4 Lexicography and Knowledge Acquisition 4.1 The Lexicon Structure 4.2 Lexicon Acquisition 4.3 A Multipurpose Processing and Acquisitino Environment 5 Source Language Analysis 5.1 The Phases of Natural Language Analysis 5.2 Analysis Algorithms 5.2.1 The Generalized LR (Tomita) Algorithm 5.3 Augmented Context-Free Grammars 5.3.1 Computational LFG for Source Language Analysis 5.3.1.1 Compiling CLFG into Tomita-Style LR Tables 5.3.2 Enhancements for KBMT 5.4 Integration of Syntactic and Semantic Analysis 5.4.1 The Universal Parser Architecture 5.5 Beyond the Universal Parser Architecture 5.6 Concluding Remarks 6 Target Language Generation 6.1 Text Planning 6.2 Text Plan Formalism 6.3 Decision Knowledge in Text Planning 6.4 Lexical Selection 6.4.1 Collocation and Synonymy 6.4.2 The Clause Level 6.4.3 Reference Treatment 6.5 Generator Architecture 6.5.1 Knowledge Sources 6.6 Syntactic Selection and Realization 7 Speech-to-Speech and Translation 7.1 Speech Recognition 7.1.1 A Historical Perspective 7.1.2 Evaluation Metrics 7.1.3 Hidden Markov Models 7.1.4 Neural Networks 7.2 Integrating Speech Recognition and Language Analysis 7.3 SpeechTrans with-f Noise-Proof LR Parsing 7.3.1 Handling Erratic Phonemes 7.3.2 An Example: Integrating GLR Parsing with Speech Recognition 7.3.3 Scoring and the Confusion Matrix 7.4 Sphinx-LR: Integrating HMMs with LR Parsing 7.4.1 The HMM-LR Method 7.4.2 The Integrated Speech-Parsing Method 7.5 Speech Translation with Neural Networks 7.5.1 Speech Recognition with Linked Predictive Neural Networks 7.5.2 Integration of LPNN and KBMT 8 Machine-Aided Translation 8.1 The TWS Configuration 8.1.1 The Basic Translator's Workstation 8.1.2 An Enhanced Translator's Workstation 8.2 Other Multilingual NLP Environments 8.2.1 Multilingual Electronic Communication Environments 8.2.2 Information Processing and Abstracting 8.2.3 A Translation Shop 9 The Future of Machine Translation 9.1 From the Mainframe to the Desktop 9.2 From Monolithic Systems to Interactive Symbiosis 9.3 From "MAT versus MT" to "M(A)T" 9.4 From Knowledge-Free to Knowledge-Based Machine Translation 9.5 From Monomedia to Multimedia Machine Translation 9.6 Transferor Interlingua? 9.7 From Standalone MT to Situated MT A KBMT Glossary Bibliography Index

「Nielsen BookData」 より

詳細情報

ページトップへ