A robust combination strategy for semantic role labeling

Lluís Màrquez, Mihai Surdeanu, Pere Comas, Jordi Turmo

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

10 Citations (Scopus)

Abstract

This paper focuses on semantic role labeling using automatically-generated syntactic information. A simple and robust strategy for system combination is presented, which allows to partially recover from input parsing errors and to significantly boost results of individual systems. This combination scheme is also very flexible since the individual systems are not required to provide any information other than their solution. Extensive experimental evaluation in the CoNLL- 2005 shared task framework supports our previous claims. The proposed architecture outperforms the best results reported in that evaluation exercise.

Original languageEnglish (US)
Title of host publicationHLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
Pages644-651
Number of pages8
StatePublished - 2005
Externally publishedYes
EventHuman Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, HLT/EMNLP 2005, Co-located with the 2005 Document Understanding Conference, DUC and the 9th International Workshop on Parsing Technologies, IWPT - Vancouver, BC, Canada
Duration: Oct 6 2005Oct 8 2005

Other

OtherHuman Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, HLT/EMNLP 2005, Co-located with the 2005 Document Understanding Conference, DUC and the 9th International Workshop on Parsing Technologies, IWPT
CountryCanada
CityVancouver, BC
Period10/6/0510/8/05

Fingerprint

Syntactics
Labeling
Semantics

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Information Systems

Cite this

Màrquez, L., Surdeanu, M., Comas, P., & Turmo, J. (2005). A robust combination strategy for semantic role labeling. In HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference (pp. 644-651)

A robust combination strategy for semantic role labeling. / Màrquez, Lluís; Surdeanu, Mihai; Comas, Pere; Turmo, Jordi.

HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. 2005. p. 644-651.

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

Màrquez, L, Surdeanu, M, Comas, P & Turmo, J 2005, A robust combination strategy for semantic role labeling. in HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. pp. 644-651, Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, HLT/EMNLP 2005, Co-located with the 2005 Document Understanding Conference, DUC and the 9th International Workshop on Parsing Technologies, IWPT, Vancouver, BC, Canada, 10/6/05.
Màrquez L, Surdeanu M, Comas P, Turmo J. A robust combination strategy for semantic role labeling. In HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. 2005. p. 644-651
Màrquez, Lluís ; Surdeanu, Mihai ; Comas, Pere ; Turmo, Jordi. / A robust combination strategy for semantic role labeling. HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. 2005. pp. 644-651
@inproceedings{5003f14893834d8e9ebd4e6571fddae6,
title = "A robust combination strategy for semantic role labeling",
abstract = "This paper focuses on semantic role labeling using automatically-generated syntactic information. A simple and robust strategy for system combination is presented, which allows to partially recover from input parsing errors and to significantly boost results of individual systems. This combination scheme is also very flexible since the individual systems are not required to provide any information other than their solution. Extensive experimental evaluation in the CoNLL- 2005 shared task framework supports our previous claims. The proposed architecture outperforms the best results reported in that evaluation exercise.",
author = "Llu{\'i}s M{\`a}rquez and Mihai Surdeanu and Pere Comas and Jordi Turmo",
year = "2005",
language = "English (US)",
pages = "644--651",
booktitle = "HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

}

TY - GEN

T1 - A robust combination strategy for semantic role labeling

AU - Màrquez, Lluís

AU - Surdeanu, Mihai

AU - Comas, Pere

AU - Turmo, Jordi

PY - 2005

Y1 - 2005

N2 - This paper focuses on semantic role labeling using automatically-generated syntactic information. A simple and robust strategy for system combination is presented, which allows to partially recover from input parsing errors and to significantly boost results of individual systems. This combination scheme is also very flexible since the individual systems are not required to provide any information other than their solution. Extensive experimental evaluation in the CoNLL- 2005 shared task framework supports our previous claims. The proposed architecture outperforms the best results reported in that evaluation exercise.

AB - This paper focuses on semantic role labeling using automatically-generated syntactic information. A simple and robust strategy for system combination is presented, which allows to partially recover from input parsing errors and to significantly boost results of individual systems. This combination scheme is also very flexible since the individual systems are not required to provide any information other than their solution. Extensive experimental evaluation in the CoNLL- 2005 shared task framework supports our previous claims. The proposed architecture outperforms the best results reported in that evaluation exercise.

UR - http://www.scopus.com/inward/record.url?scp=45749111419&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=45749111419&partnerID=8YFLogxK

M3 - Conference contribution

SP - 644

EP - 651

BT - HLT/EMNLP 2005 - Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

ER -