A word-class approach to labeling PSCFG rules for machine translation

Andreas Zollmann, Stephan Vogel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

13 Citations (Scopus)

Abstract

In this work we propose methods to label probabilistic synchronous context-free grammar (PSCFG) rules using only word tags, generated by either part-of-speech analysis or unsupervised word class induction. The proposals range from simple tag-combination schemes to a phrase clustering model that can incorporate an arbitrary number of features. Our models improve translation quality over the single generic label approach of Chiang (2005) and perform on par with the syntactically motivated approach from Zollmann and Venugopal (2006) on the NIST large Chineseto- English translation task. These results persist when using automatically learned word tags, suggesting broad applicability of our technique across diverse language pairs for which syntactic resources are not available.

Original languageEnglish
Title of host publicationACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies
Pages1-11
Number of pages11
Volume1
Publication statusPublished - 1 Dec 2011
Externally publishedYes
Event49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL-HLT 2011 - Portland, OR, United States
Duration: 19 Jun 201124 Jun 2011

Other

Other49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL-HLT 2011
CountryUnited States
CityPortland, OR
Period19/6/1124/6/11

Fingerprint

grammar
induction
language
resources
Tag
Machine Translation
Grammar
Labeling
Word Class
Syntax
Induction
Resources
Language
Part of Speech
English Translation
Speech Analysis

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Zollmann, A., & Vogel, S. (2011). A word-class approach to labeling PSCFG rules for machine translation. In ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (Vol. 1, pp. 1-11)

A word-class approach to labeling PSCFG rules for machine translation. / Zollmann, Andreas; Vogel, Stephan.

ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. Vol. 1 2011. p. 1-11.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zollmann, A & Vogel, S 2011, A word-class approach to labeling PSCFG rules for machine translation. in ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. vol. 1, pp. 1-11, 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL-HLT 2011, Portland, OR, United States, 19/6/11.
Zollmann A, Vogel S. A word-class approach to labeling PSCFG rules for machine translation. In ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. Vol. 1. 2011. p. 1-11
Zollmann, Andreas ; Vogel, Stephan. / A word-class approach to labeling PSCFG rules for machine translation. ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. Vol. 1 2011. pp. 1-11
@inproceedings{1665cb0c02e64425856edc19f014f2b8,
title = "A word-class approach to labeling PSCFG rules for machine translation",
abstract = "In this work we propose methods to label probabilistic synchronous context-free grammar (PSCFG) rules using only word tags, generated by either part-of-speech analysis or unsupervised word class induction. The proposals range from simple tag-combination schemes to a phrase clustering model that can incorporate an arbitrary number of features. Our models improve translation quality over the single generic label approach of Chiang (2005) and perform on par with the syntactically motivated approach from Zollmann and Venugopal (2006) on the NIST large Chineseto- English translation task. These results persist when using automatically learned word tags, suggesting broad applicability of our technique across diverse language pairs for which syntactic resources are not available.",
author = "Andreas Zollmann and Stephan Vogel",
year = "2011",
month = "12",
day = "1",
language = "English",
isbn = "9781932432879",
volume = "1",
pages = "1--11",
booktitle = "ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies",

}

TY - GEN

T1 - A word-class approach to labeling PSCFG rules for machine translation

AU - Zollmann, Andreas

AU - Vogel, Stephan

PY - 2011/12/1

Y1 - 2011/12/1

N2 - In this work we propose methods to label probabilistic synchronous context-free grammar (PSCFG) rules using only word tags, generated by either part-of-speech analysis or unsupervised word class induction. The proposals range from simple tag-combination schemes to a phrase clustering model that can incorporate an arbitrary number of features. Our models improve translation quality over the single generic label approach of Chiang (2005) and perform on par with the syntactically motivated approach from Zollmann and Venugopal (2006) on the NIST large Chineseto- English translation task. These results persist when using automatically learned word tags, suggesting broad applicability of our technique across diverse language pairs for which syntactic resources are not available.

AB - In this work we propose methods to label probabilistic synchronous context-free grammar (PSCFG) rules using only word tags, generated by either part-of-speech analysis or unsupervised word class induction. The proposals range from simple tag-combination schemes to a phrase clustering model that can incorporate an arbitrary number of features. Our models improve translation quality over the single generic label approach of Chiang (2005) and perform on par with the syntactically motivated approach from Zollmann and Venugopal (2006) on the NIST large Chineseto- English translation task. These results persist when using automatically learned word tags, suggesting broad applicability of our technique across diverse language pairs for which syntactic resources are not available.

UR - http://www.scopus.com/inward/record.url?scp=84859044437&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859044437&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781932432879

VL - 1

SP - 1

EP - 11

BT - ACL-HLT 2011 - Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies

ER -