Improving unsupervised dependency parsing with richer contexts and smoothing

William P. Headden; Mark Johnson; David McClosky

Improving unsupervised dependency parsing with richer contexts and smoothing

William P. Headden^*, Mark Johnson, David McClosky

^*Corresponding author for this work

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

86 Citations (Scopus)

Abstract

Unsupervised grammar induction models tend to employ relatively simple models of syntax when compared to their supervised counterparts. Traditionally, the unsupervised models have been kept simple due to tractability and data sparsity concerns. In this paper, we introduce basic valence frames and lexical information into an unsupervised dependency grammar inducer and show how this additional information can be leveraged via smoothing. Our model produces state-of-the-art results on the task of unsupervised grammar induction, improving over the best previous work by almost 10 percentage points.

Original language	English
Title of host publication	NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference
Place of Publication	Stroudsburg, PA
Publisher	Association for Computational Linguistics (ACL)
Pages	101-109
Number of pages	9
ISBN (Print)	9781932432411
Publication status	Published - Jun 2009
Externally published	Yes
Event	Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies, NAACL HLT (10th : 2009) - Boulder, United States Duration: 31 May 2009 → 5 Jun 2009

Other

Other	Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies, NAACL HLT (10th : 2009)
Country/Territory	United States
City	Boulder
Period	31/05/09 → 5/06/09

Access to Document

http://dl.acm.org/citation.cfm?id=1620769

Cite this

Headden, W. P., Johnson, M., & McClosky, D. (2009). Improving unsupervised dependency parsing with richer contexts and smoothing. In NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference (pp. 101-109). Association for Computational Linguistics (ACL). http://dl.acm.org/citation.cfm?id=1620769

Headden, William P. ; Johnson, Mark ; McClosky, David. / Improving unsupervised dependency parsing with richer contexts and smoothing. NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Stroudsburg, PA : Association for Computational Linguistics (ACL), 2009. pp. 101-109

@inproceedings{46c662c0ee28489999cce68da23d36b5,

title = "Improving unsupervised dependency parsing with richer contexts and smoothing",

abstract = "Unsupervised grammar induction models tend to employ relatively simple models of syntax when compared to their supervised counterparts. Traditionally, the unsupervised models have been kept simple due to tractability and data sparsity concerns. In this paper, we introduce basic valence frames and lexical information into an unsupervised dependency grammar inducer and show how this additional information can be leveraged via smoothing. Our model produces state-of-the-art results on the task of unsupervised grammar induction, improving over the best previous work by almost 10 percentage points.",

author = "Headden, {William P.} and Mark Johnson and David McClosky",

year = "2009",

month = jun,

language = "English",

isbn = "9781932432411",

pages = "101--109",

booktitle = "NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference",

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

note = "Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies, NAACL HLT (10th : 2009) ; Conference date: 31-05-2009 Through 05-06-2009",

}

Headden, WP, Johnson, M & McClosky, D 2009, Improving unsupervised dependency parsing with richer contexts and smoothing. in NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Association for Computational Linguistics (ACL), Stroudsburg, PA, pp. 101-109, Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies, NAACL HLT (10th : 2009), Boulder, Colorado, United States, 31/05/09. <http://dl.acm.org/citation.cfm?id=1620769>

Improving unsupervised dependency parsing with richer contexts and smoothing. / Headden, William P.; Johnson, Mark; McClosky, David.
NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Stroudsburg, PA: Association for Computational Linguistics (ACL), 2009. p. 101-109.

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

TY - GEN

T1 - Improving unsupervised dependency parsing with richer contexts and smoothing

AU - Headden, William P.

AU - Johnson, Mark

AU - McClosky, David

PY - 2009/6

Y1 - 2009/6

N2 - Unsupervised grammar induction models tend to employ relatively simple models of syntax when compared to their supervised counterparts. Traditionally, the unsupervised models have been kept simple due to tractability and data sparsity concerns. In this paper, we introduce basic valence frames and lexical information into an unsupervised dependency grammar inducer and show how this additional information can be leveraged via smoothing. Our model produces state-of-the-art results on the task of unsupervised grammar induction, improving over the best previous work by almost 10 percentage points.

AB - Unsupervised grammar induction models tend to employ relatively simple models of syntax when compared to their supervised counterparts. Traditionally, the unsupervised models have been kept simple due to tractability and data sparsity concerns. In this paper, we introduce basic valence frames and lexical information into an unsupervised dependency grammar inducer and show how this additional information can be leveraged via smoothing. Our model produces state-of-the-art results on the task of unsupervised grammar induction, improving over the best previous work by almost 10 percentage points.

UR - http://www.scopus.com/inward/record.url?scp=84863408489&partnerID=8YFLogxK

M3 - Conference proceeding contribution

AN - SCOPUS:84863408489

SN - 9781932432411

SP - 101

EP - 109

BT - NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg, PA

T2 - Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies, NAACL HLT (10th : 2009)

Y2 - 31 May 2009 through 5 June 2009

ER -

Headden WP, Johnson M, McClosky D. Improving unsupervised dependency parsing with richer contexts and smoothing. In NAACL HLT 2009 - Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Stroudsburg, PA: Association for Computational Linguistics (ACL). 2009. p. 101-109

Improving unsupervised dependency parsing with richer contexts and smoothing

Abstract

Other

Access to Document

Other files and links

Fingerprint

Cite this