Semantic role labeling using complete syntactic analysis

Mihai Surdeanu, Jordi Turmo

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

26 Citations (Scopus)

Abstract

In this paper we introduce a semantic role labeling system constructed on top of the full syntactic analysis of text. The labeling problem is modeled using a rich set of lexical, syntactic, and semantic attributes and learned using one-versus-all AdaBoost classifiers. Our results indicate that even a simple approach that assumes that each semantic argument maps into exactly one syntactic phrase obtains encouraging performance, surpassing the best system that uses partial syntax by almost 6%.

Original languageEnglish (US)
Title of host publicationCoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning
Pages221-224
Number of pages4
StatePublished - 2005
Externally publishedYes
Event9th Conference on Computational Natural Language Learning, CoNLL 2005 - Ann Arbor, MI, United States
Duration: Jun 29 2005Jun 30 2005

Other

Other9th Conference on Computational Natural Language Learning, CoNLL 2005
CountryUnited States
CityAnn Arbor, MI
Period6/29/056/30/05

Fingerprint

Syntactics
Labeling
Semantics
semantics
Adaptive boosting
syntax
Classifiers
performance

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Linguistics and Language

Cite this

Surdeanu, M., & Turmo, J. (2005). Semantic role labeling using complete syntactic analysis. In CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning (pp. 221-224)

Semantic role labeling using complete syntactic analysis. / Surdeanu, Mihai; Turmo, Jordi.

CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning. 2005. p. 221-224.

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

Surdeanu, M & Turmo, J 2005, Semantic role labeling using complete syntactic analysis. in CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning. pp. 221-224, 9th Conference on Computational Natural Language Learning, CoNLL 2005, Ann Arbor, MI, United States, 6/29/05.
Surdeanu M, Turmo J. Semantic role labeling using complete syntactic analysis. In CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning. 2005. p. 221-224
Surdeanu, Mihai ; Turmo, Jordi. / Semantic role labeling using complete syntactic analysis. CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning. 2005. pp. 221-224
@inproceedings{dddfa3ab73ab407fba08de1682721234,
title = "Semantic role labeling using complete syntactic analysis",
abstract = "In this paper we introduce a semantic role labeling system constructed on top of the full syntactic analysis of text. The labeling problem is modeled using a rich set of lexical, syntactic, and semantic attributes and learned using one-versus-all AdaBoost classifiers. Our results indicate that even a simple approach that assumes that each semantic argument maps into exactly one syntactic phrase obtains encouraging performance, surpassing the best system that uses partial syntax by almost 6{\%}.",
author = "Mihai Surdeanu and Jordi Turmo",
year = "2005",
language = "English (US)",
pages = "221--224",
booktitle = "CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning",

}

TY - GEN

T1 - Semantic role labeling using complete syntactic analysis

AU - Surdeanu, Mihai

AU - Turmo, Jordi

PY - 2005

Y1 - 2005

N2 - In this paper we introduce a semantic role labeling system constructed on top of the full syntactic analysis of text. The labeling problem is modeled using a rich set of lexical, syntactic, and semantic attributes and learned using one-versus-all AdaBoost classifiers. Our results indicate that even a simple approach that assumes that each semantic argument maps into exactly one syntactic phrase obtains encouraging performance, surpassing the best system that uses partial syntax by almost 6%.

AB - In this paper we introduce a semantic role labeling system constructed on top of the full syntactic analysis of text. The labeling problem is modeled using a rich set of lexical, syntactic, and semantic attributes and learned using one-versus-all AdaBoost classifiers. Our results indicate that even a simple approach that assumes that each semantic argument maps into exactly one syntactic phrase obtains encouraging performance, surpassing the best system that uses partial syntax by almost 6%.

UR - http://www.scopus.com/inward/record.url?scp=84862283150&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862283150&partnerID=8YFLogxK

M3 - Conference contribution

SP - 221

EP - 224

BT - CoNLL 2005 - Proceedings of the Ninth Conference on Computational Natural Language Learning

ER -