Cognate-aware morphological segmentation for multilingual neural translation
| dc.contributor | Aalto-yliopisto | fi |
| dc.contributor | Aalto University | en |
| dc.contributor.author | Grönroos, Stig-Arne | en_US |
| dc.contributor.author | Virpioja, Sami | en_US |
| dc.contributor.author | Kurimo, Mikko | en_US |
| dc.contributor.department | Dept Signal Process and Acoust | en |
| dc.contributor.groupauthor | Centre of Excellence in Computational Inference, COIN | en |
| dc.contributor.groupauthor | Speech Recognition | en |
| dc.date.accessioned | 2019-01-14T09:21:53Z | |
| dc.date.available | 2019-01-14T09:21:53Z | |
| dc.date.issued | 2018-10-31 | en_US |
| dc.description | | openaire: EC/H2020/780069/EU//MeMAD | |
| dc.description.abstract | This article describes the Aalto University entry to the WMT18 News Translation Shared Task. We participate in the multilingual subtrack with a system trained under the constrained condition to translate from English to both Finnish and Estonian. The system is based on the Transformer model. We focus on improving the consistency of morphological segmentation for words that are similar orthographically, semantically, and distributionally; such words include etymological cognates, loan words, and proper names. For this, we introduce Cognate Morfessor, a multilingual variant of the Morfessor method. We show that our approach improves the translation quality particularly for Estonian, which has less resources for training the translation model. | en |
| dc.description.version | Peer reviewed | en |
| dc.format.extent | 8 | |
| dc.format.extent | 390-397 | |
| dc.format.mimetype | application/pdf | en_US |
| dc.identifier.citation | Grönroos, S-A, Virpioja, S & Kurimo, M 2018, Cognate-aware morphological segmentation for multilingual neural translation . in Third Conference on Machine Translation (WMT18); Brussels, Belgium . Association for Computational Linguistics, pp. 390-397, Conference on Machine Translation, Brussels, Belgium, 31/10/2018 . https://doi.org/10.18653/v1/W18-64037 | en |
| dc.identifier.doi | 10.18653/v1/W18-64037 | en_US |
| dc.identifier.isbn | 978-1-948087-81-0 | |
| dc.identifier.other | PURE UUID: 66d66e8f-6c09-4763-84ad-532d0e9809e2 | en_US |
| dc.identifier.other | PURE ITEMURL: https://research.aalto.fi/en/publications/66d66e8f-6c09-4763-84ad-532d0e9809e2 | en_US |
| dc.identifier.other | PURE LINK: http://aclweb.org/anthology/W18-6410 | en_US |
| dc.identifier.other | PURE FILEURL: https://research.aalto.fi/files/30804144/WMT037_1_.pdf | en_US |
| dc.identifier.uri | https://aaltodoc.aalto.fi/handle/123456789/35969 | |
| dc.identifier.urn | URN:NBN:fi:aalto-201901141152 | |
| dc.language.iso | en | en |
| dc.relation | info:eu-repo/grantAgreement/EC/H2020/780069/EU//MeMAD | en_US |
| dc.relation.ispartof | Conference on Machine Translation | en |
| dc.relation.ispartofseries | Third Conference on Machine Translation (WMT18); Brussels, Belgium | en |
| dc.rights | openAccess | en |
| dc.subject.keyword | neural machine translation | en_US |
| dc.subject.keyword | morphology | en_US |
| dc.subject.keyword | cognate | en_US |
| dc.subject.keyword | multilingual | en_US |
| dc.title | Cognate-aware morphological segmentation for multilingual neural translation | en |
| dc.type | A4 Artikkeli konferenssijulkaisussa | fi |
| dc.type.version | publishedVersion |
