The KBMT Project
A Case Study in Knowledge-Based Machine Translation
- 1st Edition - September 25, 1991
- Editors: Kenneth Goodman, Sergei Nirenburg
- Language: English
- eBook ISBN:9 7 8 - 0 - 0 8 - 0 5 1 8 9 0 - 9
Machine translation of natural languages is one of the most complex and comprehensive applications of computational linguistics and artificial intelligence. This is especially tr… Read more
Purchase options
Institutional subscription on ScienceDirect
Request a sales quoteMachine translation of natural languages is one of the most complex and comprehensive applications of computational linguistics and artificial intelligence. This is especially true of knowledge-based machine translation (KBMT) systems, which require many knowledge resources and processing modules to carry out the necessary levels of analysis, representation and generation of meaning and form. The number of real-world problems, tasks, and solutions involved in developing any realistic-size knowledge-based machine translation system is enormous. It is thus difficult for researchers in the field to learn what a system "really does".
This book fills that need with a detailed case study of a KBMT system implemented at the Center for Machine Translation at Carnegie Mellon University. The research consists in part of the creation of a system for translation between English and Japanese. The corpora used in the project were manuals for installing and maintaining IBM personal computers (sponsorship by IBM, through its Tokyo Research Laboratory) Individual chapters describe the interlingua texts used in knowledge-based machine translation, the grammar formalism embodied in the system, the grammars and lexicons and their roles in the translation process, the process of source language analysis, an augmentation module that interactively and automatically resolves ambiguities remaining after source language analysis, and the generator, which produces target language sentences. Detailed appendices illustrate the process from analysis through generation.
This book is intended for developers, researchers and advanced students in natural language processing and computational linguistics, including all those who have an interest in machine translation and machine-aided translation.
Edited by Kenneth Goodman and Sergei Nirenburg
- Preface
1 Introduction
- 1.1 Specifications and Architecture
1.2 The KBMT-89 Approach to Machine Translation
1.3 An Overview of the System and the Book
1.4 Extensions and Prospects
2 World Knowledge and Text Meaning
- 2.1 The Concept Lexicon
- 2.1.1 Knowledge Acquisition and Maintenance
2.1.2 Ontological Postulates
2.1.3 The Domain Ontology
2.2 The Interlingua Text
- 2.2.1 Varieties of World Knowledge in a Knowledge-Based MT System
2.2.2 Integration of Discourse and Propositional Knowledge
2.2.3 Representative Classes of Discourse Knowledge
2.2.4 Combining Concept Tokens into Networks
2.2.5 Interlingua Text and the Concept of Microtheories
2.2.6 Meaning and Representation
2.2.7 A Sample ILT
3 Syntactic Theory and Processing
- 3.1 Two Types of Syntactic Structure
- 3.1.1 Constituent Structure
3.1.2 Feature Structures
3.1.3 F-Structures in Machine Translation and Linguistic Theory
3.2 Unification: Building F-Structures during Parsing
- 3.2.1 C-Structures and Their Corresponding F-Structures
3.2.2 Names of F-Structures
3.2.3 An Informal Characterization of Unification
3.2.4 Building F-Structures during Parsing
3.2.5 Constraining Grammatical Features
3.2.6 F-Structures and Mapping Rules
4 Grammars in Analysis and Generation
- 4.1
- 4.1.1 Identifying Linguistic Generalizations
4.1.2 The Scope of the Grammar
4.1.3 The Content and Structure of Feature Structures
4.1.4 Undergeneration
4.1.5 Overgeneration
4.2 Basic English Sentence Structure
- 4.2.1 Heads and Projections
4.2.2 Noun Phrases
- 4.2.2.1 NP Rules and Structures
4.2.2.2 Compound Nouns
4.2.3 Verb Phrases
- 4.2.3.1 Syntactic Subcategorization
4.2.3.2 Auxiliary Verbs
4.2.3.3 Projections of V
4.2.4 Other Constructions in English Grammar
- 4.2.4.2 Relative Clauses
4.2.4.3 Passivization
4.2.5 Unexpressed Subjects
4.3 Japanese Word Order
- 4.3.1 Japanese Morphological Rules
4.3.2 Japanese Case Markers and Grammatical Functions
4.4 Generation Grammars
- 4.4.1 Ordering of Equations within Rules
4.4.2 Rule Ordering
4.5 Bidirectionality
5 Analysis Lexicons
- 5.1 Introduction and Overview
5.2 Structure of the Lexicon Entries
5.3 Notation
5.4 English Inflectional Morphology
5.5 Lexical Mapping Rules
5.6 Closed-Class Lexical Mapping
5.7 Structural Mapping Rules
5.8 Special Mapping Rules
6 Generation Lexicons
- 6.1 Differences in Analysis and Generation Lexicons
6.2 Generation Lexicon Entries
6.3 Generation Mapping Rules
6.4 Annotated Example
7 Source Text Analysis
- 7.1 Introduction and Overview
7.2 The Syntactic Parser
7.3 The Mapping Rule Interpreter
- 7.3.1 Lexical Mapping Rules
7.3.2 Structural Mapping Rules
7.3.3 Semantic Restrictions
7.4 Sample Traces
- 7.4.1 Disambiguation
7.4.2 The Traces
8 Automatic and Interactive Augmentation
- 8.1 Motivation and Design
8.2 Format Conversion
- 8.2.1 Conversions
8.2.2 Implementing the Conversions
- 8.2.2.1 Sample Conversion
8.3 Automatic Augmentation
- 8.3.1 Multiple Anaphor Resolution Strategies
8.3.2 Integrating the Strategies
8.3.3 Implementing the Resolution Strategy
8.4 Interactive Disambiguation
- 8.4.1 Starting Up
8.4.2 The User Interface
8.4.3 Using the Augmentor
8.4.4 Example
8.5 Implementation Details
- 8.5.1 LISP Dependencies
8.5.2 Composite ILTs
8.5.3 Removing Discarded ILTs
9 Target Text Generation
- 9.1 Design of the Generation Component
9.2 The Lexical Selection Module
- 9.2.1 Types of Meanings and Types of Realizations
9.2.2 Context-Dependent Selection: Subcategorization
9.2.3 Context-Independent Selection: The Matcher
- 9.2.3.1 Single Element Fillers
9.2.3.2 Enumerated Fillers
9.2.3.3 Range Fillers
9.2.3.4 The Frame Level
9.2.3.5 Calibration
9.2.4 The Lexical Selection Algorithms
9.3 F-Structure Creation
- 9.3.1 Mapping Rules
- 9.3.1.1 Some Sample Mapping Rules
9.3.2 Mapping Rules
9.3.3 The Mapper and Its Algorithm
9.4 Syntactic Generation: GENKIT
9.5 Control
- 9.5.1 Top-Level Generation Algorithm
9.5.2 Control Knowledge
A Annotated English-Japanese Trace
- A.1 Preamble
A.2 Analysis
A.3 Augmentation
A.4 Generation
B Annotated Japanese-English Trace
- B.1 Preamble
B.2 Analysis
B.3 Augmentation
B.4 Generation
C Running the Translation System
Bibliography
Index
- No. of pages: 340
- Language: English
- Edition: 1
- Published: September 25, 1991
- Imprint: Morgan Kaufmann
- eBook ISBN: 9780080518909