Probing semantic relations : exploration and identification in specialized texts / edited by Alain Auger, Caroline Barrière.

Format:

Book

Contributor:

Auger, Alain.

Barrière, Caroline.

Series:

Benjamins current topics ; 23.

Benjamins Current Topics ; 23

Language:

English

Subjects (All):

Semantics.

Relational grammar.

Physical Description:

ix, 156 p.

Edition:

1st ed.

Place of Publication:

Amsterdam ; Philadelphia : John Benjamins Pub. Co., 2010.

Language Note:

English

Summary:

Semantic relations are at the core of any representational system, and are keys to enable the next generation of information processing systems with semantic and reasoning capabilities. Acquisition, description, and formalization of semantic relations are fundamentals in computer-based systems where natural language processing is required. Probing Semantic Relations provides a state of the art of current research trends in the area of knowledge extraction from text using linguistic patterns. First published as a Special Issue of Terminology 14:1 (2008), the current book emphasizes how definitional knowledge is conveyed by conceptual and semantic relations such as synonymy, causality, hypernymy (generic-specific), and meronymy (part-whole). Showing the difficulties and successes of pattern-based approaches, the book illustrates current and future challenges in knowledge acquisition from text. This book provides new perspectives to researchers and practitioners in terminology, knowledge engineering, natural language processing, and semantics.

Contents:

Probing Semantic Relations

Editorial page

Title page

LCC data

Table of contents

About the Authors

Probing semantic relations

Pattern-based Extraction Dimensions

Evaluation

Terminological issues

Contributions to this volume

Concluding Remarks

Notes

References

Web-based extraction of semantic relation instances for terminology work

1. Introduction

2. A closer look at KPs

3. Discovering and filtering KPs

4. Extracting and ranking instances

4.1 Reliability measures

4.2 Heuristics

5. Evaluation

5.1 Setup and purpose

5.2 Interannotator (dis)agreement

5.3 Example output

5.4 Performance of ranking schemes

5.5 Augmenting the UMLS

6. KP portability

7. Conclusion

Designing and evaluating patterns for relation acquisition from texts with Caméléon

2. The state of the art: Pattern-based knowledge identification

2.1 Pattern-based approaches: A cross-disciplinary matter

2.2 Pattern-based identification of semantic relations

3. A pattern-based method for a semantic relation identification

3.1 Background and motivations

3.2 Overview of the approach

3.3 Step 1: Project pattern definition

3.4 Step 2: Enriching the conceptual model

3.5 Towards an evaluation of Caméléon

4. How the tool supports relation extraction

4.1 Setting up a new project

4.2 Pattern design and adaptation

4.3 Pattern evaluation

4.4 Text fragment selection

4.5 Model enrichment

5. Method and Corpora for evaluating the set of knowledge patterns

5.1 Pattern design by using previous work

5.2 Varied corpora

5.3 Evaluating the patterns

6. Results

6.1 Definition patterns

6.2 Taxonomic and other patterns

7. Discussion

7.1 Pattern elaboration

7.2 What is a "generic pattern"?.

7.3 Needed skills to perform the evaluation task

7.4 Need for documenting patterns

8. Conclusion

Definitional verbal patterns for semantic relation extraction

2. The role of definitional contexts for the extraction of semantic relations

2.1 Finding semantic relations in definitions

2.2 Semantic relations and corpora

2.3 Definitional knowledge extraction

3. Definitional verbal patterns related to specific semantic relationships

3.1 Analytic verbal patterns

3.2 Functional verbal patterns

3.3 Meronymic and extensional verbal patterns

3.4 Synonymic verbal patterns

3.5 Structural classification

4. ECODE, a definitional context extraction system

4.1 Corpus

4.2 Extraction of occurrences of definitional verbal patterns

4.3 Filtering of non-relevant contexts

4.4 Identifying constitutive elements

4.5 Evaluation

5. Conclusions

Acknowledgments

Patrones léxicos para la extracción de conceptos vinculados por la relación parte-todo en español

1. Introducción

2. Las relaciones conceptuales

2.1 La relación parte-todo y su clasificación

2.2 Relaciones similares a la relación parte-todo

3. Extracción de relaciones conceptuales mediante patrones

4. Una metodología basada en corpus para detectar patrones léxicos

4.1 Corpus utilizado

4.2 Fases del trabajo empírico

5. Resultados

6. Conclusiones

Referencias

Anexo 1. Clasificación de los términos vinculados por la relación parte-todo

Anexo 2. Catálogo de patrones léxicos

Expressions of uncertainty in candidate knowledge-rich contexts

2. Background

2.1 Knowledge patterns

2.2 Expressions of uncertainty

3. Relations

4. Methodology

5. Results

5.1 Quantification

5.2 Hedging

5.3 Modal verbs

5.4 Negation.

6. Conclusions and future work

Sources of examples

Index

The series Benjamins Current Topics.

Notes:

Bibliographic Level Mode of Issuance: Monograph

Includes bibliographical references and index.

ISBN:

1-282-72161-5

9786612721618

90-272-8792-9

OCLC:

705533542