Kernel Methods for Relation Extraction

Dmitry Zelenko; Chinatsu Aone; Anthony Richardella

Kernel Methods for Relation Extraction

Dmitry Zelenko, Chinatsu Aone, Anthony Richardella

Materials Research Institute (MRI)

Research output: Contribution to conference › Paper › peer-review

112 Scopus citations

Abstract

We present an application of kernel methods to extracting relations from unstructured natural language sources. We introduce kernels defined over shallow parse representations of text, and design efficient algorithms for computing the kernels. We use the devised kernels in conjunction with Support Vector Machine and Voted Perceptron learning algorithms for the task of extracting person-affiliation and organization-location relations from text. We experimentally evaluate the proposed methods and compare them with feature-based learning algorithms, with promising results.

Original language	English (US)
Pages	71-78
Number of pages	8
State	Published - 2002
Event	7th Conference on Empirical Methods in Natural Language Processing, EMNLP 2002 - Philadelphia, United States Duration: Jul 6 2002 → Jul 7 2002

Conference

Conference	7th Conference on Empirical Methods in Natural Language Processing, EMNLP 2002
Country/Territory	United States
City	Philadelphia
Period	7/6/02 → 7/7/02

All Science Journal Classification (ASJC) codes

Computational Theory and Mathematics
Computer Science Applications
Information Systems

Cite this

@conference{f95b84b2d9674222a53b1996941ff908,

title = "Kernel Methods for Relation Extraction",

abstract = "We present an application of kernel methods to extracting relations from unstructured natural language sources. We introduce kernels defined over shallow parse representations of text, and design efficient algorithms for computing the kernels. We use the devised kernels in conjunction with Support Vector Machine and Voted Perceptron learning algorithms for the task of extracting person-affiliation and organization-location relations from text. We experimentally evaluate the proposed methods and compare them with feature-based learning algorithms, with promising results.",

author = "Dmitry Zelenko and Chinatsu Aone and Anthony Richardella",

note = "Funding Information: Our work was supported through the DARPA Evidence Extraction and Link Discovery Program. Publisher Copyright: {\textcopyright} 2002 Proceedings of the 2002 Conference on Empirical Methods in Natural Language Processing, EMNLP 2002. All rights reserved.; 7th Conference on Empirical Methods in Natural Language Processing, EMNLP 2002 ; Conference date: 06-07-2002 Through 07-07-2002",

year = "2002",

language = "English (US)",

pages = "71--78",

}

TY - CONF

T1 - Kernel Methods for Relation Extraction

AU - Zelenko, Dmitry

AU - Aone, Chinatsu

AU - Richardella, Anthony

N1 - Funding Information: Our work was supported through the DARPA Evidence Extraction and Link Discovery Program. Publisher Copyright: © 2002 Proceedings of the 2002 Conference on Empirical Methods in Natural Language Processing, EMNLP 2002. All rights reserved.

PY - 2002

Y1 - 2002

N2 - We present an application of kernel methods to extracting relations from unstructured natural language sources. We introduce kernels defined over shallow parse representations of text, and design efficient algorithms for computing the kernels. We use the devised kernels in conjunction with Support Vector Machine and Voted Perceptron learning algorithms for the task of extracting person-affiliation and organization-location relations from text. We experimentally evaluate the proposed methods and compare them with feature-based learning algorithms, with promising results.

AB - We present an application of kernel methods to extracting relations from unstructured natural language sources. We introduce kernels defined over shallow parse representations of text, and design efficient algorithms for computing the kernels. We use the devised kernels in conjunction with Support Vector Machine and Voted Perceptron learning algorithms for the task of extracting person-affiliation and organization-location relations from text. We experimentally evaluate the proposed methods and compare them with feature-based learning algorithms, with promising results.

UR - http://www.scopus.com/inward/record.url?scp=85146432130&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85146432130&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85146432130

SP - 71

EP - 78

T2 - 7th Conference on Empirical Methods in Natural Language Processing, EMNLP 2002

Y2 - 6 July 2002 through 7 July 2002

ER -

Kernel Methods for Relation Extraction

Abstract

Conference

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this