Dual decomposition for parsing with non-projective head automata

Terry Koo, Alexander M. Rush, Michael Collins, Tommi Jaakkola, David Sontag

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

Abstract

This paper introduces algorithms for non-projective parsing based on dual decomposition. We focus on parsing algorithms for non-projective head automata, a generalization of head-automata models to non-projective structures. The dual decomposition algorithms are simple and efficient, relying on standard dynamic programming and minimum spanning tree algorithms. They provably solve an LP relaxation of the non-projective parsing problem. Empirically the LP relaxation is very often tight: for many languages, exact solutions are achieved on over 98% of test sentences. The accuracy of our models is higher than previous work on a broad range of datasets.

Original languageEnglish (US)
Title of host publicationEMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
Pages1288-1298
Number of pages11
StatePublished - 2010
EventConference on Empirical Methods in Natural Language Processing, EMNLP 2010 - Cambridge, MA, United States
Duration: Oct 9 2010Oct 11 2010

Other

OtherConference on Empirical Methods in Natural Language Processing, EMNLP 2010
CountryUnited States
CityCambridge, MA
Period10/9/1010/11/10

Fingerprint

Decomposition
Dynamic programming

ASJC Scopus subject areas

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

Cite this

Koo, T., Rush, A. M., Collins, M., Jaakkola, T., & Sontag, D. (2010). Dual decomposition for parsing with non-projective head automata. In EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference (pp. 1288-1298)

Dual decomposition for parsing with non-projective head automata. / Koo, Terry; Rush, Alexander M.; Collins, Michael; Jaakkola, Tommi; Sontag, David.

EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. 2010. p. 1288-1298.

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

Koo, T, Rush, AM, Collins, M, Jaakkola, T & Sontag, D 2010, Dual decomposition for parsing with non-projective head automata. in EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. pp. 1288-1298, Conference on Empirical Methods in Natural Language Processing, EMNLP 2010, Cambridge, MA, United States, 10/9/10.
Koo T, Rush AM, Collins M, Jaakkola T, Sontag D. Dual decomposition for parsing with non-projective head automata. In EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. 2010. p. 1288-1298
Koo, Terry ; Rush, Alexander M. ; Collins, Michael ; Jaakkola, Tommi ; Sontag, David. / Dual decomposition for parsing with non-projective head automata. EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. 2010. pp. 1288-1298
@inproceedings{f1dfbf9c5f3f407eaa397138c4a323f3,
title = "Dual decomposition for parsing with non-projective head automata",
abstract = "This paper introduces algorithms for non-projective parsing based on dual decomposition. We focus on parsing algorithms for non-projective head automata, a generalization of head-automata models to non-projective structures. The dual decomposition algorithms are simple and efficient, relying on standard dynamic programming and minimum spanning tree algorithms. They provably solve an LP relaxation of the non-projective parsing problem. Empirically the LP relaxation is very often tight: for many languages, exact solutions are achieved on over 98{\%} of test sentences. The accuracy of our models is higher than previous work on a broad range of datasets.",
author = "Terry Koo and Rush, {Alexander M.} and Michael Collins and Tommi Jaakkola and David Sontag",
year = "2010",
language = "English (US)",
isbn = "1932432868",
pages = "1288--1298",
booktitle = "EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

}

TY - GEN

T1 - Dual decomposition for parsing with non-projective head automata

AU - Koo, Terry

AU - Rush, Alexander M.

AU - Collins, Michael

AU - Jaakkola, Tommi

AU - Sontag, David

PY - 2010

Y1 - 2010

N2 - This paper introduces algorithms for non-projective parsing based on dual decomposition. We focus on parsing algorithms for non-projective head automata, a generalization of head-automata models to non-projective structures. The dual decomposition algorithms are simple and efficient, relying on standard dynamic programming and minimum spanning tree algorithms. They provably solve an LP relaxation of the non-projective parsing problem. Empirically the LP relaxation is very often tight: for many languages, exact solutions are achieved on over 98% of test sentences. The accuracy of our models is higher than previous work on a broad range of datasets.

AB - This paper introduces algorithms for non-projective parsing based on dual decomposition. We focus on parsing algorithms for non-projective head automata, a generalization of head-automata models to non-projective structures. The dual decomposition algorithms are simple and efficient, relying on standard dynamic programming and minimum spanning tree algorithms. They provably solve an LP relaxation of the non-projective parsing problem. Empirically the LP relaxation is very often tight: for many languages, exact solutions are achieved on over 98% of test sentences. The accuracy of our models is higher than previous work on a broad range of datasets.

UR - http://www.scopus.com/inward/record.url?scp=80053229665&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053229665&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1932432868

SN - 9781932432862

SP - 1288

EP - 1298

BT - EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

ER -