Ensemble Models for dependency parsing: Cheap and good?

Mihai Surdeanu, Christopher D. Manning

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

38 Citations (Scopus)

Abstract

Previous work on dependency parsing used various kinds of combination models but a systematic analysis and comparison of these approaches is lacking. In this paper we implemented such a study for English dependency parsing and find several non-obvious facts: (a) the diversity of base parsers is more important than complex models for learning (e.g., stacking, supervised meta-classification), (b) approximate, linear-time re-parsing algorithms guarantee well-formed dependency trees without significant performance loss, and (c) the simplest scoring model for re-parsing (unweighted voting) performs essentially as well as other more complex models. This study proves that fast and accurate ensemble parsers can be built with minimal effort.

Original languageEnglish (US)
Title of host publicationNAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference
Pages649-652
Number of pages4
StatePublished - 2010
Externally publishedYes
Event2010 Human Language Technologies Conference ofthe North American Chapter of the Association for Computational Linguistics, NAACL HLT 2010 - Los Angeles, CA, United States
Duration: Jun 2 2010Jun 4 2010

Other

Other2010 Human Language Technologies Conference ofthe North American Chapter of the Association for Computational Linguistics, NAACL HLT 2010
CountryUnited States
CityLos Angeles, CA
Period6/2/106/4/10

Fingerprint

voting
guarantee
Parsing
Ensemble
learning
performance
Parsers
time
Voting
Scoring

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Surdeanu, M., & Manning, C. D. (2010). Ensemble Models for dependency parsing: Cheap and good? In NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference (pp. 649-652)

Ensemble Models for dependency parsing : Cheap and good? / Surdeanu, Mihai; Manning, Christopher D.

NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference. 2010. p. 649-652.

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

Surdeanu, M & Manning, CD 2010, Ensemble Models for dependency parsing: Cheap and good? in NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference. pp. 649-652, 2010 Human Language Technologies Conference ofthe North American Chapter of the Association for Computational Linguistics, NAACL HLT 2010, Los Angeles, CA, United States, 6/2/10.
Surdeanu M, Manning CD. Ensemble Models for dependency parsing: Cheap and good? In NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference. 2010. p. 649-652
Surdeanu, Mihai ; Manning, Christopher D. / Ensemble Models for dependency parsing : Cheap and good?. NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference. 2010. pp. 649-652
@inproceedings{89f13c973d934af4a2b2427b2fff68f3,
title = "Ensemble Models for dependency parsing: Cheap and good?",
abstract = "Previous work on dependency parsing used various kinds of combination models but a systematic analysis and comparison of these approaches is lacking. In this paper we implemented such a study for English dependency parsing and find several non-obvious facts: (a) the diversity of base parsers is more important than complex models for learning (e.g., stacking, supervised meta-classification), (b) approximate, linear-time re-parsing algorithms guarantee well-formed dependency trees without significant performance loss, and (c) the simplest scoring model for re-parsing (unweighted voting) performs essentially as well as other more complex models. This study proves that fast and accurate ensemble parsers can be built with minimal effort.",
author = "Mihai Surdeanu and Manning, {Christopher D.}",
year = "2010",
language = "English (US)",
isbn = "1932432655",
pages = "649--652",
booktitle = "NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference",

}

TY - GEN

T1 - Ensemble Models for dependency parsing

T2 - Cheap and good?

AU - Surdeanu, Mihai

AU - Manning, Christopher D.

PY - 2010

Y1 - 2010

N2 - Previous work on dependency parsing used various kinds of combination models but a systematic analysis and comparison of these approaches is lacking. In this paper we implemented such a study for English dependency parsing and find several non-obvious facts: (a) the diversity of base parsers is more important than complex models for learning (e.g., stacking, supervised meta-classification), (b) approximate, linear-time re-parsing algorithms guarantee well-formed dependency trees without significant performance loss, and (c) the simplest scoring model for re-parsing (unweighted voting) performs essentially as well as other more complex models. This study proves that fast and accurate ensemble parsers can be built with minimal effort.

AB - Previous work on dependency parsing used various kinds of combination models but a systematic analysis and comparison of these approaches is lacking. In this paper we implemented such a study for English dependency parsing and find several non-obvious facts: (a) the diversity of base parsers is more important than complex models for learning (e.g., stacking, supervised meta-classification), (b) approximate, linear-time re-parsing algorithms guarantee well-formed dependency trees without significant performance loss, and (c) the simplest scoring model for re-parsing (unweighted voting) performs essentially as well as other more complex models. This study proves that fast and accurate ensemble parsers can be built with minimal effort.

UR - http://www.scopus.com/inward/record.url?scp=84858390758&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858390758&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84858390758

SN - 1932432655

SN - 9781932432657

SP - 649

EP - 652

BT - NAACL HLT 2010 - Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Main Conference

ER -