Neural Machine Translation for English–Kazakh with Morphological Segmentation and Synthetic Data

Antonio Toral Ruiz; Lukas Edman; Jennifer Spenader; Galiya Yeshmagambetova

Neural Machine Translation for English–Kazakh with Morphological Segmentation and Synthetic Data

Antonio Toral Ruiz, Lukas Edman, Jennifer Spenader, Galiya Yeshmagambetova

Artificial Intelligence

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

10 Citations (Scopus)

213 Downloads (Pure)

Abstract

This paper presents the systems submitted by the University of Groningen to the English-Kazakh language pair (both translation directions) for the WMT 2019 news translation task. We explore the potential benefits of (i) morphological segmentation (both unsupervised and rule-based), given the agglutinative nature of Kazakh, (ii) data from two additional languages (Turkish and Russian), given the scarcity of English-Kazakh data and (iii) synthetic data, both for the source and for the target language. Our best sub- missions ranked second for Kazakh-English and third for English-Kazakh in terms of the BLEU automatic evaluation metric.

Original language	English
Title of host publication	Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1)
Place of Publication	Forence, Italy
Publisher	Association for Computational Linguistics (ACL)
Pages	386-392
Number of pages	7
Volume	2
Publication status	Published - 1-Aug-2019

Access to Document

Neural Machine Translation for English–Kazakh with Morphological Segmentation Final publisher's version, 212 KBLicence: Unspecified

Handle.net

Persistent link

Cite this

@inproceedings{66a779dd87734d94837f5e3e3d5dbce1,

title = "Neural Machine Translation for English–Kazakh with Morphological Segmentation and Synthetic Data",

abstract = "This paper presents the systems submitted by the University of Groningen to the English-Kazakh language pair (both translation directions) for the WMT 2019 news translation task. We explore the potential benefits of (i) morphological segmentation (both unsupervised and rule-based), given the agglutinative nature of Kazakh, (ii) data from two additional languages (Turkish and Russian), given the scarcity of English-Kazakh data and (iii) synthetic data, both for the source and for the target language. Our best sub- missions ranked second for Kazakh-English and third for English-Kazakh in terms of the BLEU automatic evaluation metric.",

author = "{Toral Ruiz}, Antonio and Lukas Edman and Jennifer Spenader and Galiya Yeshmagambetova",

year = "2019",

month = aug,

day = "1",

language = "English",

volume = "2",

pages = "386--392",

booktitle = "Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1)",

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

}

Neural Machine Translation for English–Kazakh with Morphological Segmentation and Synthetic Data. / Toral Ruiz, Antonio; Edman, Lukas; Spenader, Jennifer et al.
Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1). Vol. 2 Forence, Italy: Association for Computational Linguistics (ACL), 2019. p. 386-392.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Neural Machine Translation for English–Kazakh with Morphological Segmentation and Synthetic Data

AU - Toral Ruiz, Antonio

AU - Edman, Lukas

AU - Spenader, Jennifer

AU - Yeshmagambetova, Galiya

PY - 2019/8/1

Y1 - 2019/8/1

N2 - This paper presents the systems submitted by the University of Groningen to the English-Kazakh language pair (both translation directions) for the WMT 2019 news translation task. We explore the potential benefits of (i) morphological segmentation (both unsupervised and rule-based), given the agglutinative nature of Kazakh, (ii) data from two additional languages (Turkish and Russian), given the scarcity of English-Kazakh data and (iii) synthetic data, both for the source and for the target language. Our best sub- missions ranked second for Kazakh-English and third for English-Kazakh in terms of the BLEU automatic evaluation metric.

AB - This paper presents the systems submitted by the University of Groningen to the English-Kazakh language pair (both translation directions) for the WMT 2019 news translation task. We explore the potential benefits of (i) morphological segmentation (both unsupervised and rule-based), given the agglutinative nature of Kazakh, (ii) data from two additional languages (Turkish and Russian), given the scarcity of English-Kazakh data and (iii) synthetic data, both for the source and for the target language. Our best sub- missions ranked second for Kazakh-English and third for English-Kazakh in terms of the BLEU automatic evaluation metric.

UR - http://www.aclweb.org/anthology/W19-5343

M3 - Conference contribution

VL - 2

SP - 386

EP - 392

BT - Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1)

PB - Association for Computational Linguistics (ACL)

CY - Forence, Italy

ER -

Neural Machine Translation for English–Kazakh with Morphological Segmentation and Synthetic Data

Abstract

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this