Using universal linguistic knowledge to guide grammar induction

Tahira Naseem; Harr Chen; Regina Barzilay; Mark Johnson

Using universal linguistic knowledge to guide grammar induction

Tahira Naseem^*, Harr Chen, Regina Barzilay, Mark Johnson

^*Corresponding author for this work

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

89 Citations (Scopus)

Abstract

We present an approach to grammar induction that utilizes syntactic universals to improve dependency parsing across a range of languages. Our method uses a single set of manually-specified language-independent rules that identify syntactic dependencies between pairs of syntactic categories that commonly occur across languages. During inference of the probabilistic model, we use posterior expectation constraints to require that a minimum proportion of the dependencies we infer be instances of these rules. We also automatically refine the syntactic categories given in our coarsely tagged input. Across six languages our approach outperforms state-of-the-art unsupervised methods by a significant margin.¹

Original language	English
Title of host publication	EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
Place of Publication	Stroudsburg, PA
Publisher	Association for Computational Linguistics (ACL)
Pages	1234-1244
Number of pages	11
ISBN (Print)	1932432868, 9781932432862
Publication status	Published - 2010
Event	Conference on Empirical Methods in Natural Language Processing, EMNLP 2010 - Cambridge, MA, United States Duration: 9 Oct 2010 → 11 Oct 2010

Other

Other	Conference on Empirical Methods in Natural Language Processing, EMNLP 2010
Country/Territory	United States
City	Cambridge, MA
Period	9/10/10 → 11/10/10

Cite this

@inproceedings{e22814811fed4f049033b42667d178d0,

title = "Using universal linguistic knowledge to guide grammar induction",

abstract = "We present an approach to grammar induction that utilizes syntactic universals to improve dependency parsing across a range of languages. Our method uses a single set of manually-specified language-independent rules that identify syntactic dependencies between pairs of syntactic categories that commonly occur across languages. During inference of the probabilistic model, we use posterior expectation constraints to require that a minimum proportion of the dependencies we infer be instances of these rules. We also automatically refine the syntactic categories given in our coarsely tagged input. Across six languages our approach outperforms state-of-the-art unsupervised methods by a significant margin.1",

author = "Tahira Naseem and Harr Chen and Regina Barzilay and Mark Johnson",

year = "2010",

language = "English",

isbn = "1932432868",

pages = "1234--1244",

booktitle = "EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

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

note = "Conference on Empirical Methods in Natural Language Processing, EMNLP 2010 ; Conference date: 09-10-2010 Through 11-10-2010",

}

Naseem, T, Chen, H, Barzilay, R & Johnson, M 2010, Using universal linguistic knowledge to guide grammar induction. in EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Association for Computational Linguistics (ACL), Stroudsburg, PA, pp. 1234-1244, Conference on Empirical Methods in Natural Language Processing, EMNLP 2010, Cambridge, MA, United States, 9/10/10.

Using universal linguistic knowledge to guide grammar induction. / Naseem, Tahira; Chen, Harr; Barzilay, Regina et al.
EMNLP 2010 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Stroudsburg, PA: Association for Computational Linguistics (ACL), 2010. p. 1234-1244.

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

TY - GEN

T1 - Using universal linguistic knowledge to guide grammar induction

AU - Naseem, Tahira

AU - Chen, Harr

AU - Barzilay, Regina

AU - Johnson, Mark

PY - 2010

Y1 - 2010

N2 - We present an approach to grammar induction that utilizes syntactic universals to improve dependency parsing across a range of languages. Our method uses a single set of manually-specified language-independent rules that identify syntactic dependencies between pairs of syntactic categories that commonly occur across languages. During inference of the probabilistic model, we use posterior expectation constraints to require that a minimum proportion of the dependencies we infer be instances of these rules. We also automatically refine the syntactic categories given in our coarsely tagged input. Across six languages our approach outperforms state-of-the-art unsupervised methods by a significant margin.1

AB - We present an approach to grammar induction that utilizes syntactic universals to improve dependency parsing across a range of languages. Our method uses a single set of manually-specified language-independent rules that identify syntactic dependencies between pairs of syntactic categories that commonly occur across languages. During inference of the probabilistic model, we use posterior expectation constraints to require that a minimum proportion of the dependencies we infer be instances of these rules. We also automatically refine the syntactic categories given in our coarsely tagged input. Across six languages our approach outperforms state-of-the-art unsupervised methods by a significant margin.1

UR - http://www.scopus.com/inward/record.url?scp=80053240894&partnerID=8YFLogxK

M3 - Conference proceeding contribution

AN - SCOPUS:80053240894

SN - 1932432868

SN - 9781932432862

SP - 1234

EP - 1244

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

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg, PA

T2 - Conference on Empirical Methods in Natural Language Processing, EMNLP 2010

Y2 - 9 October 2010 through 11 October 2010

ER -

Using universal linguistic knowledge to guide grammar induction

Abstract

Other

Other files and links

Fingerprint

Cite this