A non-monotonic Arc-Eager transition system for dependency parsing

Matthew Honnibal; Yoav Goldberg; Mark Johnson

A non-monotonic Arc-Eager transition system for dependency parsing

Matthew Honnibal, Yoav Goldberg, Mark Johnson

School of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

30 Citations (Scopus)

Abstract

Previous incremental parsers have used monotonic state transitions. However, transitions can be made to revise previous decisions quite naturally, based on further information. We show that a simple adjustment to the Arc-Eager transition system to relax its monotonicity constraints can improve accuracy, so long as the training data includes examples of mistakes for the nonmonotonic transitions to repair. We evaluate the change in the context of a stateof-the-art system, and obtain a statistically significant improvement (p <0.001) on the English evaluation and 5/10 of the CoNLL languages.

Original language	English
Title of host publication	CoNLL 2013
Subtitle of host publication	Seventeenth Conference on Computational Natural Language Learning : Proceeding of the Conference
Place of Publication	Stroudsburg, PA
Publisher	Association for Computational Linguistics (ACL)
Pages	163-172
Number of pages	10
ISBN (Electronic)	9781937284701
ISBN (Print)	9781937284701
Publication status	Published - 2013
Event	Conference on Computational Natural Language Learning (17th : 2013) - Soﬁa, Bulgaria Duration: 8 Aug 2013 → 9 Aug 2013

Conference

Conference	Conference on Computational Natural Language Learning (17th : 2013)
City	Soﬁa, Bulgaria
Period	8/08/13 → 9/08/13

Access to Document

http://aclweb.org/anthology/W/W13/W13-3518.pdf

Cite this

@inproceedings{1ba7332ad4534574b3be985d5fc1037e,

title = "A non-monotonic Arc-Eager transition system for dependency parsing",

abstract = "Previous incremental parsers have used monotonic state transitions. However, transitions can be made to revise previous decisions quite naturally, based on further information. We show that a simple adjustment to the Arc-Eager transition system to relax its monotonicity constraints can improve accuracy, so long as the training data includes examples of mistakes for the nonmonotonic transitions to repair. We evaluate the change in the context of a stateof-the-art system, and obtain a statistically significant improvement (p <0.001) on the English evaluation and 5/10 of the CoNLL languages.",

author = "Matthew Honnibal and Yoav Goldberg and Mark Johnson",

year = "2013",

language = "English",

isbn = "9781937284701",

pages = "163--172",

booktitle = "CoNLL 2013",

publisher = "Association for Computational Linguistics (ACL)",

note = "Conference on Computational Natural Language Learning (17th : 2013) ; Conference date: 08-08-2013 Through 09-08-2013",

}

Honnibal, M, Goldberg, Y & Johnson, M 2013, A non-monotonic Arc-Eager transition system for dependency parsing. in CoNLL 2013: Seventeenth Conference on Computational Natural Language Learning : Proceeding of the Conference. Association for Computational Linguistics (ACL), Stroudsburg, PA, pp. 163-172, Conference on Computational Natural Language Learning (17th : 2013), Soﬁa, Bulgaria, 8/08/13. <http://aclweb.org/anthology/W/W13/W13-3518.pdf>

A non-monotonic Arc-Eager transition system for dependency parsing. / Honnibal, Matthew; Goldberg, Yoav; Johnson, Mark.
CoNLL 2013: Seventeenth Conference on Computational Natural Language Learning : Proceeding of the Conference. Stroudsburg, PA: Association for Computational Linguistics (ACL), 2013. p. 163-172.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - A non-monotonic Arc-Eager transition system for dependency parsing

AU - Honnibal, Matthew

AU - Goldberg, Yoav

AU - Johnson, Mark

PY - 2013

Y1 - 2013

N2 - Previous incremental parsers have used monotonic state transitions. However, transitions can be made to revise previous decisions quite naturally, based on further information. We show that a simple adjustment to the Arc-Eager transition system to relax its monotonicity constraints can improve accuracy, so long as the training data includes examples of mistakes for the nonmonotonic transitions to repair. We evaluate the change in the context of a stateof-the-art system, and obtain a statistically significant improvement (p <0.001) on the English evaluation and 5/10 of the CoNLL languages.

AB - Previous incremental parsers have used monotonic state transitions. However, transitions can be made to revise previous decisions quite naturally, based on further information. We show that a simple adjustment to the Arc-Eager transition system to relax its monotonicity constraints can improve accuracy, so long as the training data includes examples of mistakes for the nonmonotonic transitions to repair. We evaluate the change in the context of a stateof-the-art system, and obtain a statistically significant improvement (p <0.001) on the English evaluation and 5/10 of the CoNLL languages.

UR - http://www.scopus.com/inward/record.url?scp=84951730021&partnerID=8YFLogxK

M3 - Conference proceeding contribution

SN - 9781937284701

SP - 163

EP - 172

BT - CoNLL 2013

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg, PA

T2 - Conference on Computational Natural Language Learning (17th : 2013)

Y2 - 8 August 2013 through 9 August 2013

ER -

A non-monotonic Arc-Eager transition system for dependency parsing

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this