NEVĚŘILOVÁ, Zuzana and Marie STARÁ. Neural Tagger for Czech Language: Capturing Linguistic Phenomena in Web Corpora. In Aleš Horák, Pavel Rychlý, Adam Rambousek. Proceedings of the Thirteenth Workshop on Recent Advances in Slavonic Natural Language Processing, RASLAN 2019. Brno: Tribun EU, 2019, p. 23-32. ISBN 978-80-263-1517-9.
Other formats:   BibTeX LaTeX RIS
Basic information
Original name Neural Tagger for Czech Language: Capturing Linguistic Phenomena in Web Corpora
Authors NEVĚŘILOVÁ, Zuzana (203 Czech Republic, belonging to the institution) and Marie STARÁ (203 Czech Republic, belonging to the institution).
Edition Brno, Proceedings of the Thirteenth Workshop on Recent Advances in Slavonic Natural Language Processing, RASLAN 2019, p. 23-32, 10 pp. 2019.
Publisher Tribun EU
Other information
Original language English
Type of outcome Proceedings paper
Field of Study 10201 Computer sciences, information science, bioinformatics
Country of publisher Czech Republic
Confidentiality degree is not subject to a state or trade secret
Publication form printed version "print"
WWW URL
RIV identification code RIV/00216224:14330/19:00111625
Organization unit Faculty of Informatics
ISBN 978-80-263-1517-9
ISSN 2336-4289
UT WoS 000604899800003
Keywords in English Czech Tagger; Multi-word Expressions; Pretrained WordEmbeddings
Tags International impact, Reviewed
Changed by Changed by: Mgr. Michal Petr, učo 65024. Changed: 16/5/2022 15:20.
Abstract
We propose a new tagger for the Czech language and particu-larly for the tagset used for annotation of corpora of the TenTen family.The tagger is based on neural networks with pretrained word embed-dings. We selected the newest Czech Web corpus of the TenTen familyas training data, but we removed sentences with phenomena that wereoften annotated incorrectly. We let the tagger to learn the annotation ofthese phenomena on its own. We also experimented with the recognitionof multi-word expressions since this information can support the correcttagging.We evaluated the tagger on 6,950 sentences (84,023 tokens) from thecstenten17corpus and achieved 75.25% accuracy when compared bytags. When compared by attributes, we achieved 91.62% accuracy; theaccuracy of POS tag prediction is 96.5%.
Links
EF16_013/0001781, research and development projectName: LINDAT/CLARIN - Výzkumná infrastruktura pro jazykové technologie - rozšíření repozitáře a výpočetní kapacity
LM2015071, research and development projectName: Jazyková výzkumná infrastruktura v České republice (Acronym: LINDAT-Clarin)
Investor: Ministry of Education, Youth and Sports of the CR
PrintDisplayed: 11/10/2024 14:31