Metode in orodja za lažjo pripravo korpusov usvajanja jezika

Arhar Holdt, Špela (author), Kosem, Iztok (author), Stritar Kučuk, Mojca (author)

Prispevek predstavi težka mesta izdelave korpusov usvajanja tujega in maternega jezika, kot so transkribiranje besedil, anonimizacija, ročno označevanje in vsebinsko kategoriziranje popravkov, v nadaljevanju pa novo prosto dostopno orodje, ki ponuja rešitev za opisane metodološke izzive. Orodje, ki temelji na švedskem programu Svala, smo prilagodili za slovenščino, ga nadgradili, da omogoča delo s korpusoma Šolar in KOST, ter evalvirali s pomočjo dejanske korpusne gradnje.

• Language: Slovenian
• Keywords: slovenščina, učenje tujega jezika, učenje maternega jezika, korpusi usvajanja jezika, Svala, KOST, Šolar
• Work type: Article
• Typology: 1.16 - Independent Scientific Component Part or a Chapter in a Monograph
• Organization: FF - Faculty of Arts; FRI - Faculty of Computer and Information Science
• Year: 2022
• Number of pages: Str. 23-30
• PID: 20.500.12556/RUL-143901
• UDC: 811.163.6'243:81'322
• DOI: 10.4312/Obdobja.41.23-30
• COBISS.SI-ID: 129063939

Copyright

Licenca navedena na pristajalni strani zbornika.

• Publication date in RUL: 18.01.2023

Record is a part of a monograph

• Title: Na stičišču svetov : slovenščina kot drugi in tuji jezik
• Editors: Nataša Pirih Svetina, Ina Ferbežar
• Place of publishing: Ljubljana
• Publisher: Založba Univerze
• Year: 2022
• ISBN: 978-961-297-026-0
• COBISS.SI-ID: 128310531
• Collection title: Zbirka Obdobja
• Collection numbering: 41
• Collection ISSN: 1408-211X

Licences

• License: CC BY-NC-SA 4.0, Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
• Link: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Description: A Creative Commons license that bans commercial use and requires the user to release any modified works under this license.

Secondary language

• Language: English

Abstract

This article highlights the challenges of creating learner and developmental text corpora that feature error corrections: transcription and anonymization of texts, and manual annotation and categorization of corrections. It presents a new freely available tool that offers a solution to these challenges. Based on the Swedish Svala software, the tool has been adapted for Slovenian, modified to work with the Šolar and KOST corpora, and evaluated as part of an actual corpus creation process.

• Keywords: Slovene, foreign language learning, learner corpus, developmental corpus, Svala, KOST, Šolar

Projects

• Funder: ARRS - Slovenian Research Agency
• Project number: J7-3159-2021
• Name: Empirična podlaga za digitalno podprt razvoj pisne jezikovne zmožnosti

• Funder: ARRS - Slovenian Research Agency
• Project number: P6-0411-2019
• Name: Jezikovni viri in tehnologije za slovenski jezik