Recent advances in natural language processing III : selected papers from RANLP 2003 / edited by Nicolas Nicolov ... [et al.].

Format:

Book

Conference/Event

Author/Creator:

RANLP 2003, Corporate Author.

Contributor:

Nicolov, Nicolas.

Conference Name:

RANLP 2003 (2003 : Samokov, Bulgaria)

RANLP 2003

Series:

Amsterdam studies in the theory and history of linguistic science. Current issues in linguistic theory ; Series IV, v. 260.

Amsterdam studies in the theory and history of linguistic science. Series IV, Current issues in linguistic theory, 0304-0763 ; v. 260

Language:

English

Subjects (All):

Computational linguistics--Congresses.

Computational linguistics.

Applied linguistics--Congresses.

Applied linguistics.

Physical Description:

xii, 402 p. : ill.

Edition:

1st ed.

Place of Publication:

Amsterdam ; Philadelphia : J. Benjamins, c2004.

Language Note:

English

Summary:

This volume brings together revised versions of a selection of papers presented at the 2003 International Conference on "Recent Advances in Natural Language Processing". A wide range of topics is covered in the volume: semantics, dialogue, summarization, anaphora resolution, shallow parsing, morphology, part-of-speech tagging, named entity, question answering, word sense disambiguation, information extraction. Various 'state-of-the-art' techniques are explored: finite state processing, machine learning (support vector machines, maximum entropy, decision trees, memory-based learning, inductive logic programming, transformation-based learning, perceptions), latent semantic analysis, constraint programming. The papers address different languages (Arabic, English, German, Slavic languages) and use different linguistic frameworks (HPSG, LFG, constraint-based DCG).This book will be of interest to those who work in computational linguistics, corpus linguistics, human language technology, translation studies, cognitive science, psycholinguistics, artificial intelligence, and informatics.

Contents:

RECENT ADVANCES IN NATURAL LANGUAGE PROCESSING III

Editorial page

Title page

LCC data

Contents

Editors' Foreword

A Type-Theoretic Approach to Anaphora and Ellipsis Resolution

Abstract

1 Introduction

2 PTCT

3 An intensional number theory

4 Representing proportional generalized quantifiers in PTCT

5 A type-theoretical approach to anaphora

6 Ellipsis

7 Comparison with other type-theoretical approaches

8 Conclusions and future work

REFERENCES

Human Dialogue Modelling Using Machine Learning

2 Modality independent dialogue management

3 Learning to annotate utterances

4 Future work: Data driven dialogue discovery

5 Discussion

Acknowledgements

Learning Domain Theories

1 Domain theories

2 A partial ATIS domain theory

3 Scaling up

4 Domain theory for company succession events

5 Next steps

Recent Developments in Temporal Information Extraction

2 Temporal information extraction

3 Previous research

4 TimeML

5 TIMEX2

6 TIMEX2 tagging

7 TIMEX3 extensions

8 Challenges in TimeML link annotation

9 Empirical constraints on temporal discourse

10 Automatic TLINK tagging

11 Multilinguality

12 Conclusions

Annotation-Based Finite State Processing in a Large-Scale NLP Architecture

2 Finite-state technology and annotations

3 Pattern matching over annotations

4 A design for annotations-based FS matching

5 Conclusion

Acquiring Lexical Paraphrases from a Single Corpus

2 Background and related work

3 Algorithm

4 Evaluation and analysis

5 Conclusions

REFERENCES.

Multi-Word Collocation Extraction by Syntactic Composition of Collocation Bigrams

2 Existing methods of multi-word collocation extraction

3 Collocation bigrams extraction with FipsCo

4 Multi-word collocation extraction by bigrams composition

5 The experiment. Results and discussion

6 Conclusions and future work

Combining Independent Modules in Lexical Multiple-Choice Problems

2 Module combination

3 Synonyms

4 Analogies

Roget's Thesaurus and Semantic Similarity

2 Roget's Thesaurus relations as a measure of semantic distance

3 Evaluation based on human judgment

4 Evaluation based on synonymy problems

Clustering WordNet Word Senses

2 Retrieving examples for word senses from theWeb

3 Constructing topic signatures

4 The Cluto clustering environment

5 Clustering using WSD system confusion matrixes

6 Clustering using translation similarities

7 Clustering using word sense examples

8 Clustering using topic signatures

9 Experiments

10 Conclusions and future work

Inducing Hyperlinking Rules in Text Collections

2 Rule-based hyperlinkingwith information extraction techniques

3 Shallow discourse analysis for automatically deriving hyperlinking rules

4 Analysis of the results in a financial domain

Near-Synonym Choice in Natural Language Generation

2 Meta-concepts

3 Near-synonym choice

4 Preferences

5 Similarity of distinctions

6 Similarity of conceptual configurations

7 Evaluation of Xenon.

8 Conclusion

Acknowledgments

Fast and Accurate Part-of-Speech Tagging: The SVM Approach Revisited

2 Support Vector Machines

3 Problem setting

4 Experiments

Part-of-Speech Tagging with Minimal Lexicalization

1 Part-of-Speech Tagging

2 POS tagging Bayesian net

3 Experiments and Results

4 Conclusion

Accurate Annotation: An Efficiency Metric

2 An objective metric for annotation efficiency

3 Accurate tagging

4 Efficiency

Annex

Structured Parameter Estimation for LFG-DOP

2 Tree-DOP: Phrase-structure

3 Backoff Estimation for DOP

4 LFG-DOP: Lexical-Functional Grammar

5 BackOff Estimation for LFG-DOP

6 Conclusions

Parsing Without Grammar-Using Complete Trees Instead

2 The Tübingen Treebank of Spoken German, TüBa-D/S

3 Parsing function-argument structure using an instance base of trees

4 Evaluation

5 Conclusion and future work

Phrase Recognition by Filtering and Ranking with Perceptrons

2 The phrase recognition model

3 Online learning via recognition feedback

4 Phrase recognition in natural language

5 Feature vector representation

6 System implementation and experiments

7 Conclusions

Cascaded Finite-State Partial Parsing: A Larger-First Approach

2 Motivation

3 The Syntactic Relation Set

4 The larger-first algorithm

5 Evaluation

A Constraint-based Bottom-up Counterpart to Definite Clause Grammars

Abstract.

1 Introduction

3 Syntax, semantics and implementation of CHRG

4 Examples

5 Abduction in CHRG

6 Conclusion

Using Parallel Texts to Improve Recall in Botany

1 Overview - parallel texts and greedy extraction

2 Parallel texts - corpus and data

3 Shallow parsing

4 Evaluation of information merging

5 Greedy extraction

6 Prospects

Marking Atomic Events in Sets of Related Texts

2 A study of event annotation

3 Detecting and labeling events

4 System output

5 Comparison with Information Extraction

6 System evaluation

Semantically Driven Approach for Scenario Recognition in the IE System FRET

2 FRET Architecture

3 Logical form translation

4 Inference mechanism

6 Conclusions and further work

A Framework for Named Entity Recognition in the Open Domain

2 The method for named entity recognition in the open domain

3 The test corpus

5 Related work

Latent Semantic Analysis and the Construction of Coherent Extracts

2 Algorithm

3 Evaluation

4 Results

Facilitating Email Thread Access by Extractive Summary Generation

2 Related work

3 Task and corpus

4 Generating thread overviews

5 Evaluation and results

Towards Deeper Understanding of the Latent Semantic Analysis Performance

2 LSA and text categorisation

3 Related work and motivation.

4 Linguistic resources and text collections

5 Experiments and evaluation

6 Discussion

7 Conclusions and future work

Automatic Linking of Similar Texts Across Languages

1 Introduction and motivation

3 Mapping documents to the EUROVOC thesaurus

4 Calculation of cross-lingual document similarity

5 Application to multilingual news analysis

6 Future work

Verb Phrase Ellipsis Detection Using Machine Learning Techniques

2 Detection of elided VPEs

3 Conclusion and future work

HPSG-based Annotation Scheme for Corpora Development and Parsing Evaluation

2 Formalism for HPSG

3 Corpus annotation

4 Reclassification

5 Evaluation over an HPSG annotation scheme

6 Related work and discussion

Arabic Morpho-syntax for Text-to-Speech

1 Outline

2 Morphology

3 Syntax

4 Conclusions

Guessing Morphological Classes of Unknown German Nouns

3 Resources: lexicons and grammatical knowledge

6 Improvement by linear context

7 Conclusion and future work

Building Sense Tagged Corpora with Volunteer Contributions over the Web

2 Building sense tagged corpora with the help ofWeb users

3 Quantity and quality ofWeb-based sense tagged corpora

4 Exploiting agreement of human annotators forWSD

5 Summary

Reducing False Positives by Expert Combination in Automatic Keyword Indexing

2 Training the classifiers

3 Combining the experts

4 Conclusions and future work

Socrates: A Question Answering Prototype for Bulgarian.

Notes:

Papers from the RANLP conference held Sept. 10-12, 2003 in Samokov, Bulgaria.

Includes bibliographical references and index.

ISBN:

9786612254321

9781423766469

1423766466

9789027294685

9027294682

9781282254329

1282254324

OCLC:

70774073