Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation

Antonio Miceli Barone; Giuseppe  Attardi

Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation

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

Abstract

The quality of statistical machine translation performed with phrase based approaches can be increased by permuting the words in the source sentences in an order which resembles that of the target language. We propose a class of recurrent neural models which exploit sourceside dependency syntax features to reorder the words into a target-like order. We evaluate these models on the German to-English
language pair, showing significant improvements over a phrase-based
Moses baseline, obtaining a quality similar or superior to that of hand-coded syntactical reordering rules.

Original language	English
Title of host publication	Proceedings of SSST-9, Ninth Workshop on Syntax, Semantics and Structure in Statistical Translation
Publisher	Association for Computational Linguistics (ACL)
Pages	10-20
Number of pages	11
ISBN (Print)	978-1-941643-41-9
Publication status	Published - Jun 2015

Access to Document

W15-1002Final published version, 182 KB

Cite this

@inproceedings{cf666637dccf4e47ba5410b0180fefbb,

title = "Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation",

abstract = "The quality of statistical machine translation performed with phrase based approaches can be increased by permuting the words in the source sentences in an order which resembles that of the target language. We propose a class of recurrent neural models which exploit sourceside dependency syntax features to reorder the words into a target-like order. We evaluate these models on the German to-Englishlanguage pair, showing significant improvements over a phrase-basedMoses baseline, obtaining a quality similar or superior to that of hand-coded syntactical reordering rules.",

author = "{Miceli Barone}, Antonio and Giuseppe Attardi",

year = "2015",

month = jun,

language = "English",

isbn = "978-1-941643-41-9",

pages = "10--20",

booktitle = "Proceedings of SSST-9, Ninth Workshop on Syntax, Semantics and Structure in Statistical Translation",

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

}

Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation. / Miceli Barone, Antonio; Attardi, Giuseppe .

Proceedings of SSST-9, Ninth Workshop on Syntax, Semantics and Structure in Statistical Translation. Association for Computational Linguistics (ACL), 2015. p. 10-20.

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

TY - GEN

T1 - Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation

AU - Miceli Barone, Antonio

AU - Attardi, Giuseppe

PY - 2015/6

Y1 - 2015/6

N2 - The quality of statistical machine translation performed with phrase based approaches can be increased by permuting the words in the source sentences in an order which resembles that of the target language. We propose a class of recurrent neural models which exploit sourceside dependency syntax features to reorder the words into a target-like order. We evaluate these models on the German to-Englishlanguage pair, showing significant improvements over a phrase-basedMoses baseline, obtaining a quality similar or superior to that of hand-coded syntactical reordering rules.

AB - The quality of statistical machine translation performed with phrase based approaches can be increased by permuting the words in the source sentences in an order which resembles that of the target language. We propose a class of recurrent neural models which exploit sourceside dependency syntax features to reorder the words into a target-like order. We evaluate these models on the German to-Englishlanguage pair, showing significant improvements over a phrase-basedMoses baseline, obtaining a quality similar or superior to that of hand-coded syntactical reordering rules.

M3 - Conference contribution

SN - 978-1-941643-41-9

SP - 10

EP - 20

BT - Proceedings of SSST-9, Ninth Workshop on Syntax, Semantics and Structure in Statistical Translation

PB - Association for Computational Linguistics (ACL)

ER -

Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation

Abstract

Access to Document

Fingerprint

Cite this