TED-MWE: a bilingual parallel corpus with MWE annotation

Johanna Monti; Federico Sangati; MIHAEL ARCAN

doi:10.13025/20932

TED-MWE: a bilingual parallel corpus with MWE annotation

Johanna Monti
, Federico Sangati
, MIHAEL ARCAN

Computer Science

Research output: Chapter in Book or Conference Publication/Proceeding › Conference Publication › peer-review

Abstract

The translation of Multiword expressions (MWE) by Machine Translation (MT) represents a big challenge, and although MT has considerably improved in recent years, MWE mistranslations still occur very frequently. There is the need to develop large data sets, mainly parallel corpora, annotated with MWEs, since they are useful both for SMT training purposes and MWE translation quality evaluation. This paper describes a methodology to annotate a parallel spoken corpus with MWEs. The dataset used for this experiment is an English-Italian corpus extracted from the TED spoken corpus and complemented by an SMT output.

Original language	English (Ireland)
Title of host publication	Second Italian Conference on Computational Linguistics (CLiC-it 2015)
Publisher	Accademia University Press
DOIs	https://doi.org/10.13025/20932 https://doi.org/10.4000/books.aaccademia.1514
Publication status	Published - 1 Jan 2015

Access to Document

10.13025/20932Licence: CC BY-NC-ND
10.4000/books.aaccademia.1514

Clic15Accepted author manuscript, 225 KBLicence: CC BY-NC-ND

Cite this

@inproceedings{c65e19039a1c442096303097a585d7ce,

title = "TED-MWE: a bilingual parallel corpus with MWE annotation",

abstract = "The translation of Multiword expressions (MWE) by Machine Translation (MT) represents a big challenge, and although MT has considerably improved in recent years, MWE mistranslations still occur very frequently. There is the need to develop large data sets, mainly parallel corpora, annotated with MWEs, since they are useful both for SMT training purposes and MWE translation quality evaluation. This paper describes a methodology to annotate a parallel spoken corpus with MWEs. The dataset used for this experiment is an English-Italian corpus extracted from the TED spoken corpus and complemented by an SMT output.",

author = "Johanna Monti and Federico Sangati and MIHAEL ARCAN",

year = "2015",

month = jan,

day = "1",

doi = "10.13025/20932",

language = "English (Ireland)",

booktitle = "Second Italian Conference on Computational Linguistics (CLiC-it 2015)",

publisher = "Accademia University Press",

}

TY - GEN

T1 - TED-MWE: a bilingual parallel corpus with MWE annotation

AU - Monti, Johanna

AU - Sangati, Federico

AU - ARCAN, MIHAEL

PY - 2015/1/1

Y1 - 2015/1/1

N2 - The translation of Multiword expressions (MWE) by Machine Translation (MT) represents a big challenge, and although MT has considerably improved in recent years, MWE mistranslations still occur very frequently. There is the need to develop large data sets, mainly parallel corpora, annotated with MWEs, since they are useful both for SMT training purposes and MWE translation quality evaluation. This paper describes a methodology to annotate a parallel spoken corpus with MWEs. The dataset used for this experiment is an English-Italian corpus extracted from the TED spoken corpus and complemented by an SMT output.

AB - The translation of Multiword expressions (MWE) by Machine Translation (MT) represents a big challenge, and although MT has considerably improved in recent years, MWE mistranslations still occur very frequently. There is the need to develop large data sets, mainly parallel corpora, annotated with MWEs, since they are useful both for SMT training purposes and MWE translation quality evaluation. This paper describes a methodology to annotate a parallel spoken corpus with MWEs. The dataset used for this experiment is an English-Italian corpus extracted from the TED spoken corpus and complemented by an SMT output.

UR - http://hdl.handle.net/10379/14901

U2 - 10.13025/20932

DO - 10.13025/20932

M3 - Conference Publication

BT - Second Italian Conference on Computational Linguistics (CLiC-it 2015)

PB - Accademia University Press

ER -

TED-MWE: a bilingual parallel corpus with MWE annotation

Abstract

Access to Document

Other files and links

Fingerprint

Cite this