Recent Advances in Natural Language Processing

Name: Recent Advances in Natural Language Processing | Selected Papers from RANLP '95
Brand: John Benjamins Publishing Company
Price: 161.99 EUR
Availability: OnlineOnly

Selected Papers from RANLP '95

Ruslan Mitkov Nicolas Nicolov(Herausgeber*in)

John Benjamins Publishing Company

1. Auflage

Erschienen am 20. November 1997

XII, 472 Seiten

E-Book

PDF mit Adobe-DRM

Systemvoraussetzungen

978-90-272-7600-1 (ISBN)

161,99 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Adobe-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Personen

Inhalt

RECENT ADVANCES IN NATURAL LANGUAGE PROCESSING
Editorial page
Title page
Copyright page
Table of contents
Editors' Foreword
I. MORPHOLOGY AND SYNTAX
Some Linguistic, Computational and StatisticalImplications of Lexicalised Grammars
Abstract
1. Lexicalisation
2. Lexicalised Tree-Adjoining Grammar
3. Statistical implications
4. Synchronous TAGs
5. Viewing lexicalised trees as super parts-of-speech
6. LTAGs and explanation-based learning techniques
6.1 Implications of LTAG representation for EBL
7. LTAGs and Categorial Grammars
REFERENCES
Case and word order in English and German
Abstract
1. Background
2. Case and order in English
Subject affinity rules
Linear precedence rules
3. Case and word order in German
4. Implementation
5. English is German
REFERENCES
An Optimised Algorithm for Data Oriented Parsing
Abstract
1. Introduction
2. STSGs: Definitions, terminology and properties
3. Disambiguating an input sentence
4. Experimental results
5. Conclusions
REFERENCES
Parsing Repairs
Abstract
1. Introduction
2. Characterising self-repair
2.1 The overt characteristics of self-repair
2.2 The structural characteristics of self-repair
3. Analysing self-repair
3.1 Syntactic well-formedness
3.2 Cascaded repair
3.3 Interpreting O#R
3.4 Parsing self-repairs
4. Representing self-repair
5. An augmented Earley algorithm for repair
5.1 String representations
5.2 Augmentations to the standard algorithm
5.3 The augmented algorithm
5.4 Remarks
6 Conclusion
REFERENCES
Parsing for Targeted Errors in Controlled Languages
Abstract
1. Introduction
2. Controlled language and grammar checking
3. Robust parsing
3.1 Positive and negative detection
3.2 Targeted and untargeted detection
3.3 Single phase and multiple phase
3.4 Current methods
4. Chart parsing with finite state automata
5. Encoding grammatical variation with finite state automata
5.1 Deletion
5.2 Insertion
6. Complexity
7. Further work
8. Conclusions
REFERENCES
Applicative and Combinatory Categorial Grammar(from syntax to functional semantics)
Abstract
1. Model of Applicative and Cognitive Grammar
1.1 Categorial grammars
1.2 Applicative and Combinatory Categorial Grammar
2. Structural reorganisation
3. Coordination
4. Meta-rules
5. Examples
6. Conclusion
REFERENCES
PARSETALK about Textual Ellipsis
Abstract
1. Introduction
2. Ontological engineering for ellipsis resolution
3. Functional centering principles
4. Grammatical predicates for textual ellipsis
5. Text cohesion parsing: Ellipsis resolution
6. Comparison with related approaches
7. Conclusion
REFERENCES
Improving a Robust Morphological Analyser Using Lexical Transducers
Abstract
1. Introduction
2. Lexical transducers
3. The standard analyser
4. The analysis and correction of linguistic variants
5. The analysis of unknown words
6. Conclusions
REFERENCES
II. SEMANTICS AND DISAMBIGUATION
Context-Sensitive Word Distanceby Adaptive Scaling of a Semantic Space
Abstract
1. Introduction
2. Vector-representation of word meaning
2.1 From an English dictionary to P-vectors
2.2 From P-vectors to Q-vectors
3. Adaptive scaling of the semantic space
3.1 Semantic subspaces
3.2 Adaptive scaling
4. Examples of measuring the word distance
5. Evaluation through word prediction
6. Discussion
6.1 Semantic vectors
6.2 Word prediction and text structure
7. Conclusion
REFERENCES
Towards a Sublanguage-BasedSemantic Clustering Algorithm
Abstract
1. Introduction
2. Epsilon [?]: Knowledge acquisition as an evolutionary process
3. Knowledge acquisition process
3.1 POS information
3.2 Modular configuration
4. The Compound subprocess
4.1 Framework
5. The Classify subprocess
5.1 Inverse KWIC
5.2 Evaluation
6. Central knowledge base
7. Dynamic context matching techniques for semantic clusteringdisambiguation
7.1 Word sense disambiguation
7.2 Dynamic context matching
8. Concluding remarks
REFERENCES
Customising a Verb Classification to a Sublanguage
Abstract
1. Sense disambiguation and sense tuning
2. A context-based classifier
3. Discussion of the results
4. Final remarks
REFERENCES
Concept-Driven Search AlgorithmIncorporating Semantic Interpretationand Speech Recognition
Abstract
1. Introduction
2. Semantic interpretation based on concepts
2.1 Basic process
2.2 Reduction of ambiguity in concept hypotheses
3. Integrating speech recognition
3.1 Basic process
3.2 Speech understanding experiments
4. Improving search efficiency
4.1 Basic process
4.2 Speech understanding experiments
5. Concluding remarks
REFERENCES
A Proposal for Word Sense DisambiguationUsing Conceptual Distance
Abstract
1. Introduction
2. WordNet and the semantic concordance
3. Conceptual density and word sense disambiguation
4. The disambiguation algorithm using conceptual density
5. The experiment
6. Conclusions
REFERENCES
An Episodic Memory for Understanding and Learning
Abstract
1. Introduction
2. Structure of the episodic memory
2.1 Text representation
2.2 The episodic memory
3. Episode matching and memorisation
3.1 Similarity of TUs
3.2 Similarity of slots and similarity of graphs
3.3 Memorisation of an episode: The aggregation process
4. Conclusion
REFERENCES
Ambiguities & Ambiguity Labelling:Towards Ambiguity Data Bases
Abstract
1. Introduction
2. A formal view of ambiguities
2.1 Levels and contexts of ambiguities
2.1.1 Three levels of granularity for ambiguity labelling
2.1.2 Task-derived limitations on utterance-level ambiguities
2.1.3 Necessity to consider utterance-level ambiguities in the context offull utterances
2.2 Representation systems
2.2.1 Types of formal representation systems
2.2.2 Computable representations and 'reasonable' analysers
2.2.3 Expectations for a system of manual labelling
2.3 Ambiguous representations
2.3.1 Proper representations
2.4 Scope, occurrence, kernel and type of ambiguity
2.4.1 Informal presentation
2.4.2 Scope of an ambiguity
2.4.3 Occurrence and kernel of an ambiguity
2.4.4 Ambiguity type and ambiguity pattern
3. Attributes and values used in manual labelling
3.1 Top level (piece)
3.2 Paragraph or turn level
3.2.1 Structure of the list and associated separators
3.2.2 Representation of ambiguities of segmentation
3.3 Utterance level
3.3.1 Structure of the lists and associated separators
3.3.2 Headers of ambiguity kernels
3.3.3 Obligatory labels
3.3.4 Other labels
4. Conclusions
REFERENCES
III. DISCOURSE
Incorporating Discourse Aspectsin English - Polish MT
Abstract
1. Introduction
2. Centering model for English analysis
2.1 Extension to the centering algorithm
3. Local discourse mechanisms in translation
4. Ordering of Polish constituents
4.1 Ordering criteria
4.2 Building on orders of constituents
5. Conclusion
REFERENCES
Two Engines Are Better Than One:Generating More Power and Confidencein the Search for the Antecedent
Abstract
1. Introduction
2. An integrated anaphor resolution approach
3. An uncertainty reasoning approach
4. The two-engine strategy
5. Illustration
6. Conclusion
REFERENCES
Effects of Grammatical Annotation on a TopicIdentification Task
Abstract
1. Introduction
2. Topic recognition model
3. Text representation
4. Experiments
4.1 Test setting
4.2 Result and analysis
5. Conclusion
REFERENCES
Discourse Constraints on Theme Selection
Abstract
1. Introduction
2. Text type, subject matter and theme selection
3. Theme selection as interstratal constraints
4. Conclusions
REFERENCES
Discerning Relevant Information in Discourses Using TFA
Abstract
1. Introduction
2. The communication of information
3. Rhetorical structure of turns
4. An example
5. Discerning relevant information
6. Conclusions
REFERENCES
IV. GENERATION
Approximate Chart Generationfrom Non-Hierarchical Representations
Abstract
1. Introduction
2. Generation from non-hierarchical representations
3. D-Tree Grammars
4. Knowledge sources
4.1 Mapping rules
5. Sentence generation
5.1 Building a skeletal structure
5.2 Covering the remaining semantics
5.3 Completing a derivation
6. Example
7. Matching the applicability semantics of mapping rules
8. Preference-based chart generation
9. Implementation
10. Discussion
11. Conclusion
REFERENCES
Example-Based Optimisationof Surface-Generation Tables
Abstract
1. Introduction
2. LR compilation for parsing
3. The semantic head-driven generation algorithm
4. Grammar inversion
5. LR compilation for generation
6. The generation algorithm
7. Optimising the generation tables
8. An example-based optimisation technique
9. Discussion
REFERENCES
Sentence Generation by Pattern Matching:The Problem of Syntactic Choice
Abstract
1. Introduction: The speaker's problem
2. What kind of evidence can we provide in favour of patternmatching?
3. Why cognitive linguistics, or, why study natural language inthe realm of cognitive science?
4. Where do linguistic structures come from?
5. Conceptual structures and syntactic structures are to a greatextent parallel
5.1 Discussion
6. What do syntactic structures depend upon?
7. Prototypical patterns
8. Where do relative clauses come from, how can they be recognised,and what do they depend upon?
8.1 Discussion
9. Discussion
10. Conclusion
REFERENCES
An Empirical Study on the Generation ofDescriptions for Nominal Anaphors in Chinese
Abstract
1. Introduction
2. Analysis of nominal anaphors in the test data
3. A preference rule for nominal descriptions
4. Experimental results
5. Implementation
6. Conclusion
REFERENCES
Generation of Multilingual Explanations from Conceptual Graphs
Abstract
1. Introduction
2. Conceptual graphs: A brief introduction
3. Our internal representation
4. Relevant system components
5. Multilingual generation of explanations
5.1 The main objectives - subject information, coherence and multilinguality
5.2 Query mapper - the strategical component
5.3 EGEN - the tactical component
5.3.1 Input
5.3.2 Explanation levels
5.3.3 Utterance forming
5.4 Sample output
6. Implementation
7. Conclusion
REFERENCES
V. CORPUS PROCESSING ANDAPPLICATIONS
Machine Translation: Productivity and Conventionality of Language
Abstract
1. Introduction
2. Disappointment
3. Myth-1: Compositionality of translation
4. Myth-2: Possible translation
5. Examples: Metonymic nature of language and translation
6. Conceptual design of a simple MT system
7. Other frameworks and future directions
REFERENCES
Connectionist F-structure Transfer
Abstract
1. Introduction
2. F-structure representations
3. The mapper
4. An example
5. Training, testing and performance
6. Discussion
7. Conclusion
REFERENCES
Acquisition of Translation Rulesfrom Parallel Corpora
Abstract
1. Introduction
2. Acquisition of Translation Rules
2.1 Calculation of Word Similarities
2.2 Structural matching of parallel sentences
2.3 Acquisition of translation rules
3. Experiments of translation rule acquisition
3.1 Acquisition of translation rules
3.2 The translation rules
4. Discussion and Related Works
5. Conclusions
REFERENCES
Clause Recognition in the Framework of Alignment
Abstract
1. Introduction
2. Previous work
3. The model
4. Results
5. Conclusions
REFERENCES
Bilingual Vocabulary Estimationfrom Noisy Parallel CorporaUsing Variable Bag Estimation
Abstract
1. Introduction
2. Related work
3. Methodology
3.1 General estimation of distribution
3.2 Variable bag estimation
4. Experiments
5. Results and observations
6. Conclusions
REFERENCES
A HMM Part-of-Speech Tagger for Koreanwith Wordphrasal Relations
Abstract
1. Introduction
2. Wordphrase based Hidden Markov Model
2.1 Characteristics of Korean
2.2 Construction of Hidden Markov Model
3. Experiments
3.1 Test data
3.2 Results
4. Conclusions
REFERENCES
A Multimodal Environment for TelecommunicationSpecifications
Abstract
1. Introduction
2. System overview
3. Automata and rules
4. The visual language
5. Modality synergy
6. Architecture of the NL component
7. A simple translation example
8. Conclusions
REFERENCES
List and Addresses of Contributors
Index of Subjects and Terms

Systemvoraussetzungen

Als PDF speichern Als Link merken