Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging

Long Duong; Paul Cook; Steven Bird; Pavel Pecina

Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging

Long Duong, Paul Cook, Steven Bird, Pavel Pecina

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper published in Proceedings › peer-review

Abstract

Bilingual corpora offer a promising bridge between resource-rich and resource-poor languages, enabling the development of natural language processing systems for the latter. English is often selected as the resource-rich language, but another choice might give better performance. In this paper, we consider the task of unsupervised cross-lingual POS tagging, and construct a model that predicts the best source language for a given target language. In experiments on 9 languages, this model improves on using a single fixed source language. We then show that further improvements can be made by combining information from multiple source languages.

Original language	English
Title of host publication	Proceedings of the Sixth International Joint Conference on Natural Language Processing
Pages	1243-1249
Number of pages	7
Publication status	Published - 2013
Externally published	Yes
Event	International Joint Conference on Natural Language Processing - Nagoya, Japan Duration: 14 Oct 2013 → 18 Oct 2013 Conference number: 6th

Conference

Conference	International Joint Conference on Natural Language Processing
Country/Territory	Japan
City	Nagoya
Period	14/10/13 → 18/10/13

Access to Document

https://www.aclweb.org/anthology/I13-1177Licence: CC BY

Cite this

@inproceedings{e8e2bf1f4b63424fb4137517734cef83,

title = "Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging",

abstract = "Bilingual corpora offer a promising bridge between resource-rich and resource-poor languages, enabling the development of natural language processing systems for the latter. English is often selected as the resource-rich language, but another choice might give better performance. In this paper, we consider the task of unsupervised cross-lingual POS tagging, and construct a model that predicts the best source language for a given target language. In experiments on 9 languages, this model improves on using a single fixed source language. We then show that further improvements can be made by combining information from multiple source languages.",

author = "Long Duong and Paul Cook and Steven Bird and Pavel Pecina",

year = "2013",

language = "English",

pages = "1243--1249",

booktitle = "Proceedings of the Sixth International Joint Conference on Natural Language Processing",

note = "International Joint Conference on Natural Language Processing ; Conference date: 14-10-2013 Through 18-10-2013",

}

Duong, L, Cook, P, Bird, S & Pecina, P 2013, Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging. in Proceedings of the Sixth International Joint Conference on Natural Language Processing. pp. 1243-1249, International Joint Conference on Natural Language Processing, Nagoya, Japan, 14/10/13. <https://www.aclweb.org/anthology/I13-1177>

Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging. / Duong, Long; Cook, Paul; Bird, Steven et al.
Proceedings of the Sixth International Joint Conference on Natural Language Processing. 2013. p. 1243-1249.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper published in Proceedings › peer-review

TY - GEN

T1 - Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging

AU - Duong, Long

AU - Cook, Paul

AU - Bird, Steven

AU - Pecina, Pavel

N1 - Conference code: 6th

PY - 2013

Y1 - 2013

N2 - Bilingual corpora offer a promising bridge between resource-rich and resource-poor languages, enabling the development of natural language processing systems for the latter. English is often selected as the resource-rich language, but another choice might give better performance. In this paper, we consider the task of unsupervised cross-lingual POS tagging, and construct a model that predicts the best source language for a given target language. In experiments on 9 languages, this model improves on using a single fixed source language. We then show that further improvements can be made by combining information from multiple source languages.

AB - Bilingual corpora offer a promising bridge between resource-rich and resource-poor languages, enabling the development of natural language processing systems for the latter. English is often selected as the resource-rich language, but another choice might give better performance. In this paper, we consider the task of unsupervised cross-lingual POS tagging, and construct a model that predicts the best source language for a given target language. In experiments on 9 languages, this model improves on using a single fixed source language. We then show that further improvements can be made by combining information from multiple source languages.

M3 - Conference Paper published in Proceedings

SP - 1243

EP - 1249

BT - Proceedings of the Sixth International Joint Conference on Natural Language Processing

T2 - International Joint Conference on Natural Language Processing

Y2 - 14 October 2013 through 18 October 2013

ER -

Increasing the quality and quantity of source language data for unsupervised cross-lingual POS tagging

Abstract

Conference

Access to Document

Fingerprint

Cite this