Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation

Chenggang Mi; Shaolin Zhu; Yi Fan; Lei Xie

doi:10.1007/978-3-030-85896-4_27

Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation

Chenggang Mi, Shaolin Zhu, Yi Fan, Lei Xie

School of Computer Science

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

2 Scopus citations

Abstract

In this paper, we propose to use rich semantic and typological information of languages to improve the language selection method for multilingual NMT. In particular, we first use a graph-based model to output the most semantic similarity languages; then, a random forest model is built which integrates features such as data size, language family, word formation, morpheme overlap, word order, POS tag and syntax similarity together to predict the final target language(s). Experimental results on several datasets show that our method achieves consistent improvements over existing approaches both on language selection and multilingual NMT.

Original language	English
Title of host publication	Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings
Editors	Leong Hou U, Marc Spaniol, Yasushi Sakurai, Junying Chen
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	348-357
Number of pages	10
ISBN (Print)	9783030858957
DOIs	https://doi.org/10.1007/978-3-030-85896-4_27
State	Published - 2021
Event	5th International Joint Conference on Asia-Paciﬁc Web and Web-Age Information Management, APWeb-WAIM 2021 - Guangzhou, China Duration: 23 Aug 2021 → 25 Aug 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12858 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Joint Conference on Asia-Paciﬁc Web and Web-Age Information Management, APWeb-WAIM 2021
Country/Territory	China
City	Guangzhou
Period	23/08/21 → 25/08/21

Keywords

Language selection
Neural machine translation
Typological feature

Access to Document

10.1007/978-3-030-85896-4_27

Cite this

Mi, C., Zhu, S., Fan, Y., & Xie, L. (2021). Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation. In L. H. U, M. Spaniol, Y. Sakurai, & J. Chen (Eds.), Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings (pp. 348-357). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12858 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-85896-4_27

Mi, Chenggang ; Zhu, Shaolin ; Fan, Yi et al. / Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation. Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings. editor / Leong Hou U ; Marc Spaniol ; Yasushi Sakurai ; Junying Chen. Springer Science and Business Media Deutschland GmbH, 2021. pp. 348-357 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{32a00cf8d7144097907781c9d1b910d8,

title = "Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation",

abstract = "In this paper, we propose to use rich semantic and typological information of languages to improve the language selection method for multilingual NMT. In particular, we first use a graph-based model to output the most semantic similarity languages; then, a random forest model is built which integrates features such as data size, language family, word formation, morpheme overlap, word order, POS tag and syntax similarity together to predict the final target language(s). Experimental results on several datasets show that our method achieves consistent improvements over existing approaches both on language selection and multilingual NMT.",

keywords = "Language selection, Neural machine translation, Typological feature",

author = "Chenggang Mi and Shaolin Zhu and Yi Fan and Lei Xie",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 5th International Joint Conference on Asia-Paciﬁc Web and Web-Age Information Management, APWeb-WAIM 2021 ; Conference date: 23-08-2021 Through 25-08-2021",

year = "2021",

doi = "10.1007/978-3-030-85896-4_27",

language = "英语",

isbn = "9783030858957",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "348--357",

editor = "U, {Leong Hou} and Marc Spaniol and Yasushi Sakurai and Junying Chen",

booktitle = "Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings",

}

Mi, C, Zhu, S, Fan, Y & Xie, L 2021, Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation. in LH U, M Spaniol, Y Sakurai & J Chen (eds), Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12858 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 348-357, 5th International Joint Conference on Asia-Paciﬁc Web and Web-Age Information Management, APWeb-WAIM 2021, Guangzhou, China, 23/08/21. https://doi.org/10.1007/978-3-030-85896-4_27

Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation. / Mi, Chenggang; Zhu, Shaolin; Fan, Yi et al.
Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings. ed. / Leong Hou U; Marc Spaniol; Yasushi Sakurai; Junying Chen. Springer Science and Business Media Deutschland GmbH, 2021. p. 348-357 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12858 LNCS).

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

TY - GEN

T1 - Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation

AU - Mi, Chenggang

AU - Zhu, Shaolin

AU - Fan, Yi

AU - Xie, Lei

PY - 2021

Y1 - 2021

N2 - In this paper, we propose to use rich semantic and typological information of languages to improve the language selection method for multilingual NMT. In particular, we first use a graph-based model to output the most semantic similarity languages; then, a random forest model is built which integrates features such as data size, language family, word formation, morpheme overlap, word order, POS tag and syntax similarity together to predict the final target language(s). Experimental results on several datasets show that our method achieves consistent improvements over existing approaches both on language selection and multilingual NMT.

AB - In this paper, we propose to use rich semantic and typological information of languages to improve the language selection method for multilingual NMT. In particular, we first use a graph-based model to output the most semantic similarity languages; then, a random forest model is built which integrates features such as data size, language family, word formation, morpheme overlap, word order, POS tag and syntax similarity together to predict the final target language(s). Experimental results on several datasets show that our method achieves consistent improvements over existing approaches both on language selection and multilingual NMT.

KW - Language selection

KW - Neural machine translation

KW - Typological feature

UR - http://www.scopus.com/inward/record.url?scp=85115190602&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-85896-4_27

DO - 10.1007/978-3-030-85896-4_27

M3 - 会议稿件

AN - SCOPUS:85115190602

SN - 9783030858957

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 348

EP - 357

BT - Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings

A2 - U, Leong Hou

A2 - Spaniol, Marc

A2 - Sakurai, Yasushi

A2 - Chen, Junying

PB - Springer Science and Business Media Deutschland GmbH

T2 - 5th International Joint Conference on Asia-Paciﬁc Web and Web-Age Information Management, APWeb-WAIM 2021

Y2 - 23 August 2021 through 25 August 2021

ER -

Mi C, Zhu S, Fan Y, Xie L. Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation. In U LH, Spaniol M, Sakurai Y, Chen J, editors, Web and Big Data - 5th International Joint Conference, APWeb-WAIM 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 348-357. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-85896-4_27

Incorporating Typological Features into Language Selection for Multilingual Neural Machine Translation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this