<?xml version="1.0"?><rdf:RDF xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:edm="http://www.europeana.eu/schemas/edm/" xmlns:wgs84_pos="http://www.w3.org/2003/01/geo/wgs84_pos" xmlns:foaf="http://xmlns.com/foaf/0.1/" xmlns:rdaGr2="http://rdvocab.info/ElementsGr2" xmlns:oai="http://www.openarchives.org/OAI/2.0/" xmlns:owl="http://www.w3.org/2002/07/owl#" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:ore="http://www.openarchives.org/ore/terms/" xmlns:skos="http://www.w3.org/2004/02/skos/core#" xmlns:dcterms="http://purl.org/dc/terms/"><edm:WebResource rdf:about="http://www.dlib.si/stream/URN:NBN:SI:doc-I9685A02/e038c2e7-c7ad-4686-9f5f-9df0b59189b5/PDF"><dcterms:extent>283 KB</dcterms:extent></edm:WebResource><edm:TimeSpan rdf:about="2005-2025"><edm:begin xml:lang="en">2005</edm:begin><edm:end xml:lang="en">2025</edm:end></edm:TimeSpan><edm:ProvidedCHO rdf:about="URN:NBN:SI:doc-I9685A02"><dcterms:isPartOf rdf:resource="https://www.dlib.si/details/URN:NBN:SI:spr-M10OPZYO" /><dcterms:issued>2025</dcterms:issued><dc:creator>Arhar Holdt, Špela</dc:creator><dc:creator>Božič, Martin</dc:creator><dc:creator>Klemen, Matej</dc:creator><dc:creator>Robnik Šikonja, Marko</dc:creator><dc:format xml:lang="sl">številka:3</dc:format><dc:format xml:lang="sl">letnik:76</dc:format><dc:format xml:lang="sl">str. 162–176</dc:format><dc:identifier>ISSN:0038-0474</dc:identifier><dc:identifier>DOI:10.63384/sptB53z793a</dc:identifier><dc:identifier>COBISSID_HOST:259208195</dc:identifier><dc:identifier>URN:URN:NBN:SI:doc-I9685A02</dc:identifier><dc:language>en</dc:language><dc:publisher xml:lang="sl">Zveza društev pedagoških delavcev Slovenije</dc:publisher><dcterms:isPartOf xml:lang="sl">Sodobna pedagogika</dcterms:isPartOf><dc:subject xml:lang="en">educational domain</dc:subject><dc:subject xml:lang="en">grammatical error correction</dc:subject><dc:subject xml:lang="sl">izobraževalna domena</dc:subject><dc:subject xml:lang="en">large language models</dc:subject><dc:subject xml:lang="sl">popravljanje slovničnih napak</dc:subject><dc:subject xml:lang="sl">sintetiziranje podatkov</dc:subject><dc:subject xml:lang="en">synthetic data construction</dc:subject><dc:subject xml:lang="sl">veliki jezikovni modeli</dc:subject><dcterms:temporal rdf:resource="2005-2025" /><dc:title xml:lang="sl">Grammatical error correction of Slovenian school essays using large language models|</dc:title><dc:description xml:lang="sl">Grammatical error correction (GEC) is the task of automatically detecting and correcting grammatical errors in text. Large language models have enabled the development of accurate automated methods for detecting and correcting certain types of errors. In the educational domain, the aim of GEC is to aid teachers in correcting student errors. Excessive paraphrasing is a property of Generative Pre-trained Transformer-based models and is undesirable in the language education context. To avoid this, we develop multiple Slovenian models for correcting errors in spelling, word case (capitalization), word form, and word order. We describe the training data construction, training process, and model evaluation approach using the Šolar-Eval 1.0 corpus of school essays authored by primary and secondary school students. Our quantitative evaluation shows that the developed models have reasonably high accuracy levels, and our qualitative evaluation highlights the strengths and weaknesses of the models and the evaluation process. The analysis reveals multiple challenges and promising future directions for improving both model development and the evaluation process</dc:description><dc:description xml:lang="sl">Strojno popravljanje slovničnih napak je naloga, ki zajema samodejno zaznavanje in popravljanje slovničnih napak v besedilu. Na področju izobraževanja je cilj metod pomagati učiteljem pri popravljanju napak učencev. Veliki jezikovni modeli omogočajo razvoj natančnih avtomatskih metod za zaznavanje in popravljanje določenih vrst napak. Da bi se izognili pretiranemu parafraziranju, ki je značilno za modele tipa GPT, in je v kontekstu poučevanja jezika nezaželeno, predstavimo več razvitih slovenskih modelov tipa BERT in T5 za popravljanje različnih vrst napak. Te vključujejo črkovalne napake, napake v rabi velikih začetnic, besednih oblik in besednega reda. V članku opišemo postopek ustvarjanja učnih podatkov, postopek učenja ter postopek evalvacije modelov na korpusu Šolar-Eval 1.0, ki vsebuje šolske spise osnovnošolcev in srednješolcev. Avtomatska evalvacija kaže razmeroma visoko natančnost razvitih modelov, medtem ko ročna kvalitativna evalvacija razkrije prednosti in slabosti razvitih modelov ter evalvacijskega postopka. Analiza razkriva številne izzive in obetavne smeri za nadaljnje izboljšave tako pri razvoju modelov kot pri postopku evalvacije</dc:description><edm:type>TEXT</edm:type><dc:type xml:lang="sl">znanstveno časopisje</dc:type><dc:type xml:lang="en">journals</dc:type><dc:type rdf:resource="http://www.wikidata.org/entity/Q361785" /></edm:ProvidedCHO><ore:Aggregation rdf:about="http://www.dlib.si/?URN=URN:NBN:SI:doc-I9685A02"><edm:aggregatedCHO rdf:resource="URN:NBN:SI:doc-I9685A02" /><edm:isShownBy rdf:resource="http://www.dlib.si/stream/URN:NBN:SI:doc-I9685A02/e038c2e7-c7ad-4686-9f5f-9df0b59189b5/PDF" /><edm:rights rdf:resource="http://creativecommons.org/licenses/by-sa/4.0/" /><edm:provider>Slovenian National E-content Aggregator</edm:provider><edm:intermediateProvider xml:lang="en">National and University Library of Slovenia</edm:intermediateProvider><edm:dataProvider xml:lang="sl">Zveza društev pedagoških delavcev Slovenije</edm:dataProvider><edm:object rdf:resource="http://www.dlib.si/streamdb/URN:NBN:SI:doc-I9685A02/maxi/edm" /><edm:isShownAt rdf:resource="http://www.dlib.si/details/URN:NBN:SI:doc-I9685A02" /></ore:Aggregation></rdf:RDF>