Machine Translation for English--Inuktitut with Segmentation, Data Acquisition and Pre-Training

Christian Roest; Lukas Edman; Gosse Minnema; Kevin Kelly; Jennifer Spenader; Antonio Toral

Machine Translation for English--Inuktitut with Segmentation, Data Acquisition and Pre-Training

Christian Roest, Lukas Edman, Gosse Minnema, Kevin Kelly, Jennifer Spenader, Antonio Toral

Artificial Intelligence

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

24 Downloads (Pure)

Abstract

Translating to and from low-resource polysynthetic languages present numerous challenges for NMT. We present the results of our systems for the English--Inuktitut language pair for the WMT 2020 translation tasks. We investigated the importance of correct morphological segmentation, whether or not adding data from a related language (Greenlandic) helps, and whether using contextual word embeddings improves translation. While each method showed some promise, the results are mixed.

Original language	English
Title of host publication	Proceedings of the Fifth Conference on Machine Translation (WMT)
Publisher	Association for Computational Linguistics (ACL)
Pages	274-281
Number of pages	8
Publication status	Published - Nov-2020
Event	Fifth Conference on Machine Translation - Online Duration: 19-Nov-2020 → 20-Nov-2020

Conference

Conference	Fifth Conference on Machine Translation
Abbreviated title	WMT20
Period	19/11/2020 → 20/11/2020

Access to Document

Machine Translation for English–Inuktitut with Segmentation, Data Acquisition and Pre-TrainingFinal publisher's version, 200 KBLicence: CC BY

Cite this

@inproceedings{ce2469633b304064ab65ae9e5e506c5e,

title = "Machine Translation for English--Inuktitut with Segmentation, Data Acquisition and Pre-Training",

abstract = "Translating to and from low-resource polysynthetic languages present numerous challenges for NMT. We present the results of our systems for the English--Inuktitut language pair for the WMT 2020 translation tasks. We investigated the importance of correct morphological segmentation, whether or not adding data from a related language (Greenlandic) helps, and whether using contextual word embeddings improves translation. While each method showed some promise, the results are mixed.",

author = "Christian Roest and Lukas Edman and Gosse Minnema and Kevin Kelly and Jennifer Spenader and Antonio Toral",

year = "2020",

month = nov,

language = "English",

pages = "274--281",

booktitle = "Proceedings of the Fifth Conference on Machine Translation (WMT)",

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

note = "Fifth Conference on Machine Translation, WMT20 ; Conference date: 19-11-2020 Through 20-11-2020",

}

Machine Translation for English--Inuktitut with Segmentation, Data Acquisition and Pre-Training. / Roest, Christian; Edman, Lukas; Minnema, Gosse; Kelly, Kevin; Spenader, Jennifer ; Toral, Antonio.

Proceedings of the Fifth Conference on Machine Translation (WMT). Association for Computational Linguistics (ACL), 2020. p. 274-281.

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

TY - GEN

T1 - Machine Translation for English--Inuktitut with Segmentation, Data Acquisition and Pre-Training

AU - Roest, Christian

AU - Edman, Lukas

AU - Minnema, Gosse

AU - Kelly, Kevin

AU - Spenader, Jennifer

AU - Toral, Antonio

PY - 2020/11

Y1 - 2020/11

N2 - Translating to and from low-resource polysynthetic languages present numerous challenges for NMT. We present the results of our systems for the English--Inuktitut language pair for the WMT 2020 translation tasks. We investigated the importance of correct morphological segmentation, whether or not adding data from a related language (Greenlandic) helps, and whether using contextual word embeddings improves translation. While each method showed some promise, the results are mixed.

AB - Translating to and from low-resource polysynthetic languages present numerous challenges for NMT. We present the results of our systems for the English--Inuktitut language pair for the WMT 2020 translation tasks. We investigated the importance of correct morphological segmentation, whether or not adding data from a related language (Greenlandic) helps, and whether using contextual word embeddings improves translation. While each method showed some promise, the results are mixed.

M3 - Conference contribution

SP - 274

EP - 281

BT - Proceedings of the Fifth Conference on Machine Translation (WMT)

PB - Association for Computational Linguistics (ACL)

T2 - Fifth Conference on Machine Translation

Y2 - 19 November 2020 through 20 November 2020

ER -