A Bayesian LDA-based model for semi-supervised part-of-speech tagging

Kristina Toutanova; Mark Johnson

A Bayesian LDA-based model for semi-supervised part-of-speech tagging

Kristina Toutanova^*, Mark Johnson

^*Corresponding author for this work

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

34 Citations (Scopus)

Abstract

We present a novel Bayesian model for semi-supervised part-of-speech tagging. Our model extends the Latent Dirichlet Allocation model and incorporates the intuition that words' distributions over tags, p(t|w), are sparse. In addition we introduce a model for determining the set of possible tags of a word which captures important dependencies in the ambiguity classes of words. Our model outperforms the best previously proposed model for this task on a standard dataset.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference
Editors	John C. Platt, Daphne Koller, Yoram Singer, Sam T. Roweis
Place of Publication	La Jolla, California
Pages	1464-1471
Number of pages	8
Publication status	Published - 2009
Externally published	Yes
Event	21st Annual Conference on Neural Information Processing Systems, NIPS 2007 - Vancouver, BC, Canada Duration: 3 Dec 2007 → 6 Dec 2007

Other

Other	21st Annual Conference on Neural Information Processing Systems, NIPS 2007
Country/Territory	Canada
City	Vancouver, BC
Period	3/12/07 → 6/12/07

Cite this

@inproceedings{17ed2c1f25bd4c0e937cf3a0e61dd32d,

title = "A Bayesian LDA-based model for semi-supervised part-of-speech tagging",

abstract = "We present a novel Bayesian model for semi-supervised part-of-speech tagging. Our model extends the Latent Dirichlet Allocation model and incorporates the intuition that words' distributions over tags, p(t|w), are sparse. In addition we introduce a model for determining the set of possible tags of a word which captures important dependencies in the ambiguity classes of words. Our model outperforms the best previously proposed model for this task on a standard dataset.",

author = "Kristina Toutanova and Mark Johnson",

year = "2009",

language = "English",

isbn = "160560352X",

pages = "1464--1471",

editor = "Platt, {John C.} and Daphne Koller and Yoram Singer and Roweis, {Sam T.}",

booktitle = "Advances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference",

note = "21st Annual Conference on Neural Information Processing Systems, NIPS 2007 ; Conference date: 03-12-2007 Through 06-12-2007",

}

Toutanova, K & Johnson, M 2009, A Bayesian LDA-based model for semi-supervised part-of-speech tagging. in JC Platt, D Koller, Y Singer & ST Roweis (eds), Advances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference. La Jolla, California, pp. 1464-1471, 21st Annual Conference on Neural Information Processing Systems, NIPS 2007, Vancouver, BC, Canada, 3/12/07.

A Bayesian LDA-based model for semi-supervised part-of-speech tagging. / Toutanova, Kristina; Johnson, Mark.
Advances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference. ed. / John C. Platt; Daphne Koller; Yoram Singer; Sam T. Roweis. La Jolla, California, 2009. p. 1464-1471.

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

TY - GEN

T1 - A Bayesian LDA-based model for semi-supervised part-of-speech tagging

AU - Toutanova, Kristina

AU - Johnson, Mark

PY - 2009

Y1 - 2009

N2 - We present a novel Bayesian model for semi-supervised part-of-speech tagging. Our model extends the Latent Dirichlet Allocation model and incorporates the intuition that words' distributions over tags, p(t|w), are sparse. In addition we introduce a model for determining the set of possible tags of a word which captures important dependencies in the ambiguity classes of words. Our model outperforms the best previously proposed model for this task on a standard dataset.

AB - We present a novel Bayesian model for semi-supervised part-of-speech tagging. Our model extends the Latent Dirichlet Allocation model and incorporates the intuition that words' distributions over tags, p(t|w), are sparse. In addition we introduce a model for determining the set of possible tags of a word which captures important dependencies in the ambiguity classes of words. Our model outperforms the best previously proposed model for this task on a standard dataset.

UR - http://www.scopus.com/inward/record.url?scp=84863410099&partnerID=8YFLogxK

M3 - Conference proceeding contribution

AN - SCOPUS:84863410099

SN - 160560352X

SN - 9781605603520

SP - 1464

EP - 1471

BT - Advances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference

A2 - Platt, John C.

A2 - Koller, Daphne

A2 - Singer, Yoram

A2 - Roweis, Sam T.

CY - La Jolla, California

T2 - 21st Annual Conference on Neural Information Processing Systems, NIPS 2007

Y2 - 3 December 2007 through 6 December 2007

ER -

A Bayesian LDA-based model for semi-supervised part-of-speech tagging

Abstract

Other

Other files and links

Cite this