A new subtree-transfer approach to syntax-based reordering for statistical machine translation

Maxim Khalilov, José A R Fonollosa, Mark Dras

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionResearchpeer-review

Abstract

In this paper we address the problem of translating between languages with word order disparity. The idea of augmenting statistical machine translation (SMT) by using a syntax-based reordering step prior to translation, proposed in recent years, has been quite successful in improving translation quality. We present a new technique for extracting syntax-based reordering rules, which are derived through a syntactically augmented alignment of source and target texts. The parallel corpus with reordered source side is then passed to an N-gram-based machine translation system and the obtained results are contrasted with a monotone system performance. In experiments, we show significant improvement for the Chinese-to-English translation task.

LanguageEnglish
Title of host publicationProceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009
EditorsLluíz Màrquez, Harold Somers
Place of PublicationBarcelona, Spain
PublisherUniversitat Polit`ecnica de Catalunya
Pages197-204
Number of pages8
ISBN (Print)9788469239438
Publication statusPublished - 2009
Event13th Annual Conference of the European Association for Machine Translation, EAMT 2009 - Barcelona, Spain
Duration: 14 May 200915 May 2009

Other

Other13th Annual Conference of the European Association for Machine Translation, EAMT 2009
CountrySpain
CityBarcelona
Period14/05/0915/05/09

Fingerprint

Experiments
Statistical Machine Translation
Syntax
Experiment
Alignment
English Translation
Machine Translation System
Translating
Parallel Corpora
Language
N-gram

Cite this

Khalilov, M., Fonollosa, J. A. R., & Dras, M. (2009). A new subtree-transfer approach to syntax-based reordering for statistical machine translation. In L. Màrquez, & H. Somers (Eds.), Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009 (pp. 197-204). Barcelona, Spain: Universitat Polit`ecnica de Catalunya.
Khalilov, Maxim ; Fonollosa, José A R ; Dras, Mark. / A new subtree-transfer approach to syntax-based reordering for statistical machine translation. Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009. editor / Lluíz Màrquez ; Harold Somers. Barcelona, Spain : Universitat Polit`ecnica de Catalunya, 2009. pp. 197-204
@inproceedings{fe9bf831f9404e0db9b2066619bca49b,
title = "A new subtree-transfer approach to syntax-based reordering for statistical machine translation",
abstract = "In this paper we address the problem of translating between languages with word order disparity. The idea of augmenting statistical machine translation (SMT) by using a syntax-based reordering step prior to translation, proposed in recent years, has been quite successful in improving translation quality. We present a new technique for extracting syntax-based reordering rules, which are derived through a syntactically augmented alignment of source and target texts. The parallel corpus with reordered source side is then passed to an N-gram-based machine translation system and the obtained results are contrasted with a monotone system performance. In experiments, we show significant improvement for the Chinese-to-English translation task.",
author = "Maxim Khalilov and Fonollosa, {Jos{\'e} A R} and Mark Dras",
year = "2009",
language = "English",
isbn = "9788469239438",
pages = "197--204",
editor = "Llu{\'i}z M{\`a}rquez and Harold Somers",
booktitle = "Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009",
publisher = "Universitat Polit`ecnica de Catalunya",

}

Khalilov, M, Fonollosa, JAR & Dras, M 2009, A new subtree-transfer approach to syntax-based reordering for statistical machine translation. in L Màrquez & H Somers (eds), Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009. Universitat Polit`ecnica de Catalunya, Barcelona, Spain, pp. 197-204, 13th Annual Conference of the European Association for Machine Translation, EAMT 2009, Barcelona, Spain, 14/05/09.

A new subtree-transfer approach to syntax-based reordering for statistical machine translation. / Khalilov, Maxim; Fonollosa, José A R; Dras, Mark.

Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009. ed. / Lluíz Màrquez; Harold Somers. Barcelona, Spain : Universitat Polit`ecnica de Catalunya, 2009. p. 197-204.

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionResearchpeer-review

TY - GEN

T1 - A new subtree-transfer approach to syntax-based reordering for statistical machine translation

AU - Khalilov, Maxim

AU - Fonollosa, José A R

AU - Dras, Mark

PY - 2009

Y1 - 2009

N2 - In this paper we address the problem of translating between languages with word order disparity. The idea of augmenting statistical machine translation (SMT) by using a syntax-based reordering step prior to translation, proposed in recent years, has been quite successful in improving translation quality. We present a new technique for extracting syntax-based reordering rules, which are derived through a syntactically augmented alignment of source and target texts. The parallel corpus with reordered source side is then passed to an N-gram-based machine translation system and the obtained results are contrasted with a monotone system performance. In experiments, we show significant improvement for the Chinese-to-English translation task.

AB - In this paper we address the problem of translating between languages with word order disparity. The idea of augmenting statistical machine translation (SMT) by using a syntax-based reordering step prior to translation, proposed in recent years, has been quite successful in improving translation quality. We present a new technique for extracting syntax-based reordering rules, which are derived through a syntactically augmented alignment of source and target texts. The parallel corpus with reordered source side is then passed to an N-gram-based machine translation system and the obtained results are contrasted with a monotone system performance. In experiments, we show significant improvement for the Chinese-to-English translation task.

UR - http://www.scopus.com/inward/record.url?scp=84857512680&partnerID=8YFLogxK

M3 - Conference proceeding contribution

SN - 9788469239438

SP - 197

EP - 204

BT - Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009

A2 - Màrquez, Lluíz

A2 - Somers, Harold

PB - Universitat Polit`ecnica de Catalunya

CY - Barcelona, Spain

ER -

Khalilov M, Fonollosa JAR, Dras M. A new subtree-transfer approach to syntax-based reordering for statistical machine translation. In Màrquez L, Somers H, editors, Proceedings of the 13th Annual Conference of the European Association for Machine Translation, EAMT 2009. Barcelona, Spain: Universitat Polit`ecnica de Catalunya. 2009. p. 197-204