Extracting relations with integrated information using kernel methods

Shubin Zhao, Ralph Grishman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Entity relation detection is a form of information extraction that finds predefined relations between pairs of entities in text. This paper describes a relation detection approach that combines clues from different levels of syntactic processing using kernel methods. Information from three different levels of processing is considered: tokenization, sentence parsing and deep dependency analysis. Each source of information is represented by kernel functions. Then composite kernels are developed to integrate and extend individual kernels so that processing errors occurring at one level can be overcome by information from other levels. We present an evaluation of these methods on the 2004 ACE relation detection task, using Support Vector Machines, and show that each level of syntactic processing contributes useful information for this task. When evaluated on the official test data, our approach produced very competitive ACE value scores. We also compare the SVM with KNN on different kernels.

Original languageEnglish (US)
Title of host publicationACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference
Pages419-426
Number of pages8
StatePublished - 2005
Event43rd Annual Meeting of the Association for Computational Linguistics, ACL-05 - Ann Arbor, MI, United States
Duration: Jun 25 2005Jun 30 2005

Other

Other43rd Annual Meeting of the Association for Computational Linguistics, ACL-05
CountryUnited States
CityAnn Arbor, MI
Period6/25/056/30/05

Fingerprint

source of information
Kernel
Kernel Methods
evaluation
Values
Entity
Syntactic Processing
Levels of Processing
Evaluation
Sentence Parsing
Information Extraction
Support Vector Machine

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Zhao, S., & Grishman, R. (2005). Extracting relations with integrated information using kernel methods. In ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (pp. 419-426)

Extracting relations with integrated information using kernel methods. / Zhao, Shubin; Grishman, Ralph.

ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. 2005. p. 419-426.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhao, S & Grishman, R 2005, Extracting relations with integrated information using kernel methods. in ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. pp. 419-426, 43rd Annual Meeting of the Association for Computational Linguistics, ACL-05, Ann Arbor, MI, United States, 6/25/05.
Zhao S, Grishman R. Extracting relations with integrated information using kernel methods. In ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. 2005. p. 419-426
Zhao, Shubin ; Grishman, Ralph. / Extracting relations with integrated information using kernel methods. ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. 2005. pp. 419-426
@inproceedings{f8bf176f4c2247879105b03467f8e4cc,
title = "Extracting relations with integrated information using kernel methods",
abstract = "Entity relation detection is a form of information extraction that finds predefined relations between pairs of entities in text. This paper describes a relation detection approach that combines clues from different levels of syntactic processing using kernel methods. Information from three different levels of processing is considered: tokenization, sentence parsing and deep dependency analysis. Each source of information is represented by kernel functions. Then composite kernels are developed to integrate and extend individual kernels so that processing errors occurring at one level can be overcome by information from other levels. We present an evaluation of these methods on the 2004 ACE relation detection task, using Support Vector Machines, and show that each level of syntactic processing contributes useful information for this task. When evaluated on the official test data, our approach produced very competitive ACE value scores. We also compare the SVM with KNN on different kernels.",
author = "Shubin Zhao and Ralph Grishman",
year = "2005",
language = "English (US)",
isbn = "1932432515",
pages = "419--426",
booktitle = "ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference",

}

TY - GEN

T1 - Extracting relations with integrated information using kernel methods

AU - Zhao, Shubin

AU - Grishman, Ralph

PY - 2005

Y1 - 2005

N2 - Entity relation detection is a form of information extraction that finds predefined relations between pairs of entities in text. This paper describes a relation detection approach that combines clues from different levels of syntactic processing using kernel methods. Information from three different levels of processing is considered: tokenization, sentence parsing and deep dependency analysis. Each source of information is represented by kernel functions. Then composite kernels are developed to integrate and extend individual kernels so that processing errors occurring at one level can be overcome by information from other levels. We present an evaluation of these methods on the 2004 ACE relation detection task, using Support Vector Machines, and show that each level of syntactic processing contributes useful information for this task. When evaluated on the official test data, our approach produced very competitive ACE value scores. We also compare the SVM with KNN on different kernels.

AB - Entity relation detection is a form of information extraction that finds predefined relations between pairs of entities in text. This paper describes a relation detection approach that combines clues from different levels of syntactic processing using kernel methods. Information from three different levels of processing is considered: tokenization, sentence parsing and deep dependency analysis. Each source of information is represented by kernel functions. Then composite kernels are developed to integrate and extend individual kernels so that processing errors occurring at one level can be overcome by information from other levels. We present an evaluation of these methods on the 2004 ACE relation detection task, using Support Vector Machines, and show that each level of syntactic processing contributes useful information for this task. When evaluated on the official test data, our approach produced very competitive ACE value scores. We also compare the SVM with KNN on different kernels.

UR - http://www.scopus.com/inward/record.url?scp=84859917836&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859917836&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1932432515

SN - 9781932432510

SP - 419

EP - 426

BT - ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference

ER -