Turku Enhanced Parser Pipeline: From Raw Text to Enhanced Graphs in the IWPT 2020 Shared Task

Tietueen suppeat tiedot
dc.contributor.authorJenna Kanerva
dc.contributor.authorFilip Ginter
dc.contributor.authorSampo Pyysalo
dc.contributor.organizationfi=kieli- ja puheteknologia|en=Language and Speech Technology|
dc.contributor.organization-code1.2.246.10.2458963.20.47465613983
dc.converis.publication-id50410770
dc.converis.urlhttps://research.utu.fi/converis/portal/Publication/50410770
dc.date.accessioned2022-10-28T13:20:52Z
dc.date.available2022-10-28T13:20:52Z
dc.description.abstract<p>We present the approach of the TurkuNLP group to the IWPT 2020 shared task on Multilingual Parsing into Enhanced Universal Dependencies. The task involves 28 treebanks in 17 different languages and requires parsers to generate graph structures extending on the basic dependency trees. Our approach combines language-specific BERT models, the UDify parser, neural sequence-to-sequence lemmatization and a graph transformation approach encoding the enhanced structure into a dependency tree. Our submission averaged 84.5% ELAS, ranking first in the shared task.<br /></p>
dc.format.pagerange162
dc.format.pagerange173
dc.identifier.isbn978-1-952148-11-8
dc.identifier.jour-issn0736-587X
dc.identifier.olddbid181434
dc.identifier.oldhandle10024/164528
dc.identifier.urihttps://www.utupub.fi/handle/11111/46325
dc.identifier.urlhttps://www.aclweb.org/anthology/2020.iwpt-1.17
dc.identifier.urnURN:NBN:fi-fe2021042826562
dc.language.isoen
dc.okm.affiliatedauthorKanerva, Jenna
dc.okm.affiliatedauthorGinter, Filip
dc.okm.affiliatedauthorPyysalo, Sampo
dc.okm.discipline113 Computer and information sciencesen_GB
dc.okm.discipline113 Tietojenkäsittely ja informaatiotieteetfi_FI
dc.okm.internationalcopublicationnot an international co-publication
dc.okm.internationalityInternational publication
dc.okm.typeA4 Conference Article
dc.publisher.countryUnited Statesen_GB
dc.publisher.countryYhdysvallat (USA)fi_FI
dc.publisher.country-codeUS
dc.relation.conferenceInternational Conference on Parsing Technologies
dc.relation.doi10.18653/v1/2020.iwpt-1.17
dc.relation.ispartofjournalAnnual Meeting of the Association for Computational Linguistics
dc.source.identifierhttps://www.utupub.fi/handle/10024/164528
dc.titleTurku Enhanced Parser Pipeline: From Raw Text to Enhanced Graphs in the IWPT 2020 Shared Task
dc.title.bookProceedings of the 16th International Conference on Parsing Technologies and the IWPT 2020 Shared Task on Parsing into Enhanced Universal Dependencies
dc.year.issued2020

Tiedostot

Näytetään 1 - 1 / 1
Name:
2020.iwpt-1.17.pdf
Size:
346.84 KB
Format:
Adobe Portable Document Format
Description:
Publisher's PDF
Lataa

Kokoelmat

Rinnakkaistallenteet