Named Entity Recognition through Classifier Combination

Radu Florian; Abe Ittycheriah; Hongyan Jing; Tong Zhang

Named Entity Recognition through Classifier Combination

Radu Florian
, Abe Ittycheriah
, Hongyan Jing
, Tong Zhang

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

Abstract

This paper presents a classifier-combination experimental framework for named entity recognition in which four diverse classifiers (robust linear classifier, maximum entropy, transformation-based learning, and hidden Markov model) are combined under different conditions. When no gazetteer or other additional training resources are used, the combined system attains a performance of 91.6F on the English development data; integrating name, location and person gazetteers, and named entity systems trained on additional, more general, data reduces the F-measure error by a factor of 15 to 21% on the English data.

Original language	English (US)
Pages	168-171
Number of pages	4
State	Published - 2003
Externally published	Yes
Event	7th Conference on Natural Language Learning, CoNLL 2003 at HLT-NAACL 2003 - Edmonton, Canada Duration: May 31 2003 → Jun 1 2003

Conference

Conference	7th Conference on Natural Language Learning, CoNLL 2003 at HLT-NAACL 2003
Country/Territory	Canada
City	Edmonton
Period	5/31/03 → 6/1/03

ASJC Scopus subject areas

Management Science and Operations Research
Computer Graphics and Computer-Aided Design
Computer Vision and Pattern Recognition
Modeling and Simulation

Library availability

Discover UIUC Full Text

Cite this

@conference{5c5420a767184740b2583343401af54a,

title = "Named Entity Recognition through Classifier Combination",

abstract = "This paper presents a classifier-combination experimental framework for named entity recognition in which four diverse classifiers (robust linear classifier, maximum entropy, transformation-based learning, and hidden Markov model) are combined under different conditions. When no gazetteer or other additional training resources are used, the combined system attains a performance of 91.6F on the English development data; integrating name, location and person gazetteers, and named entity systems trained on additional, more general, data reduces the F-measure error by a factor of 15 to 21\% on the English data.",

author = "Radu Florian and Abe Ittycheriah and Hongyan Jing and Tong Zhang",

note = "Publisher Copyright: {\textcopyright} 2003 Proceedings of the 7th Conference on Natural Language Learning, CoNLL 2003 at HLT-NAACL 2003. All rights reserved.; 7th Conference on Natural Language Learning, CoNLL 2003 at HLT-NAACL 2003 ; Conference date: 31-05-2003 Through 01-06-2003",

year = "2003",

language = "English (US)",

pages = "168--171",

}

TY - CONF

T1 - Named Entity Recognition through Classifier Combination

AU - Florian, Radu

AU - Ittycheriah, Abe

AU - Jing, Hongyan

AU - Zhang, Tong

PY - 2003

Y1 - 2003

N2 - This paper presents a classifier-combination experimental framework for named entity recognition in which four diverse classifiers (robust linear classifier, maximum entropy, transformation-based learning, and hidden Markov model) are combined under different conditions. When no gazetteer or other additional training resources are used, the combined system attains a performance of 91.6F on the English development data; integrating name, location and person gazetteers, and named entity systems trained on additional, more general, data reduces the F-measure error by a factor of 15 to 21% on the English data.

AB - This paper presents a classifier-combination experimental framework for named entity recognition in which four diverse classifiers (robust linear classifier, maximum entropy, transformation-based learning, and hidden Markov model) are combined under different conditions. When no gazetteer or other additional training resources are used, the combined system attains a performance of 91.6F on the English development data; integrating name, location and person gazetteers, and named entity systems trained on additional, more general, data reduces the F-measure error by a factor of 15 to 21% on the English data.

UR - https://www.scopus.com/pages/publications/85122279702

UR - https://www.scopus.com/pages/publications/85122279702#tab=citedBy

M3 - Paper

AN - SCOPUS:85122279702

SP - 168

EP - 171

T2 - 7th Conference on Natural Language Learning, CoNLL 2003 at HLT-NAACL 2003

Y2 - 31 May 2003 through 1 June 2003

ER -

Named Entity Recognition through Classifier Combination

Abstract

Conference

ASJC Scopus subject areas

Library availability

Related links

Fingerprint

Cite this