BTG-based machine translation with simple reordering model using structured perceptron

Hao Wang, Yves Lepage

Research output: Contribution to conferencePaper

Abstract

In this paper, we present a novel statistical machine translation method which employs a BTG-based reordering model during decoding. BTG-based reordering models for preordering have been widely explored, aiming to improve the standard phrase-based statistical machine translation system. Less attention has been paid to incorporating such a reordering model into decoding directly. Our reordering model differs from previous models built using a syntactic parser or directly from annotated treebanks. Here, we train without using any syntactic information. The experiment results on an English-Japanese translation task show that our BTG-based decoder achieves comparable or better performance than the more complex state-of-the-art SMT decoders.

Original languageEnglish
Pages114-123
Number of pages10
Publication statusPublished - 2019 Jan 1
Event31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017 - Cebu City, Philippines
Duration: 2017 Nov 162017 Nov 18

Conference

Conference31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017
CountryPhilippines
CityCebu City
Period17/11/1617/11/18

Fingerprint

Neural networks
Syntactics
Decoding
Surface mount technology
Machine Translation
Experiments
Statistical Machine Translation
Syntax

ASJC Scopus subject areas

  • Language and Linguistics
  • Computer Science (miscellaneous)

Cite this

Wang, H., & Lepage, Y. (2019). BTG-based machine translation with simple reordering model using structured perceptron. 114-123. Paper presented at 31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017, Cebu City, Philippines.

BTG-based machine translation with simple reordering model using structured perceptron. / Wang, Hao; Lepage, Yves.

2019. 114-123 Paper presented at 31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017, Cebu City, Philippines.

Research output: Contribution to conferencePaper

Wang, H & Lepage, Y 2019, 'BTG-based machine translation with simple reordering model using structured perceptron', Paper presented at 31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017, Cebu City, Philippines, 17/11/16 - 17/11/18 pp. 114-123.
Wang H, Lepage Y. BTG-based machine translation with simple reordering model using structured perceptron. 2019. Paper presented at 31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017, Cebu City, Philippines.
Wang, Hao ; Lepage, Yves. / BTG-based machine translation with simple reordering model using structured perceptron. Paper presented at 31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017, Cebu City, Philippines.10 p.
@conference{5d2cfd4b633f4eaf8c5c3d332249a4a4,
title = "BTG-based machine translation with simple reordering model using structured perceptron",
abstract = "In this paper, we present a novel statistical machine translation method which employs a BTG-based reordering model during decoding. BTG-based reordering models for preordering have been widely explored, aiming to improve the standard phrase-based statistical machine translation system. Less attention has been paid to incorporating such a reordering model into decoding directly. Our reordering model differs from previous models built using a syntactic parser or directly from annotated treebanks. Here, we train without using any syntactic information. The experiment results on an English-Japanese translation task show that our BTG-based decoder achieves comparable or better performance than the more complex state-of-the-art SMT decoders.",
author = "Hao Wang and Yves Lepage",
year = "2019",
month = "1",
day = "1",
language = "English",
pages = "114--123",
note = "31st Pacific Asia Conference on Language, Information and Computation, PACLIC 2017 ; Conference date: 16-11-2017 Through 18-11-2017",

}

TY - CONF

T1 - BTG-based machine translation with simple reordering model using structured perceptron

AU - Wang, Hao

AU - Lepage, Yves

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we present a novel statistical machine translation method which employs a BTG-based reordering model during decoding. BTG-based reordering models for preordering have been widely explored, aiming to improve the standard phrase-based statistical machine translation system. Less attention has been paid to incorporating such a reordering model into decoding directly. Our reordering model differs from previous models built using a syntactic parser or directly from annotated treebanks. Here, we train without using any syntactic information. The experiment results on an English-Japanese translation task show that our BTG-based decoder achieves comparable or better performance than the more complex state-of-the-art SMT decoders.

AB - In this paper, we present a novel statistical machine translation method which employs a BTG-based reordering model during decoding. BTG-based reordering models for preordering have been widely explored, aiming to improve the standard phrase-based statistical machine translation system. Less attention has been paid to incorporating such a reordering model into decoding directly. Our reordering model differs from previous models built using a syntactic parser or directly from annotated treebanks. Here, we train without using any syntactic information. The experiment results on an English-Japanese translation task show that our BTG-based decoder achieves comparable or better performance than the more complex state-of-the-art SMT decoders.

UR - http://www.scopus.com/inward/record.url?scp=85072794360&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072794360&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85072794360

SP - 114

EP - 123

ER -