prikaz prve stranice dokumenta Neuronsko strojno prevođenje
  Download
PDF 1.63 MB
master's thesis
Neuronsko strojno prevođenje
Rijeka: University of Rijeka, Faculty of Humanities and Social Sciences, 2021. urn:nbn:hr:186:436094
Lerga, Košuta Estera
University of Rijeka
Faculty of Humanities and Social Sciences
Department of Italian Language and Literature
University of Rijeka
Faculty of Informatics and Digital Technologies

Institutional repository: Repository of the University of Rijeka, Faculty of Humanities and Social Sciences

Cite this document

Lerga, K. E. (2021). Neuronsko strojno prevođenje (Master's thesis). Rijeka: University of Rijeka, Faculty of Humanities and Social Sciences; Rijeka: University of Rijeka. Retrieved from https://urn.nsk.hr/urn:nbn:hr:186:436094

Lerga, Košuta Estera. "Neuronsko strojno prevođenje." Master's thesis, University of Rijeka, Faculty of Humanities and Social Sciences; University of Rijeka, 2021. https://urn.nsk.hr/urn:nbn:hr:186:436094

Lerga, Košuta Estera. "Neuronsko strojno prevođenje." Master's thesis, University of Rijeka, Faculty of Humanities and Social Sciences; University of Rijeka, 2021. https://urn.nsk.hr/urn:nbn:hr:186:436094

Lerga, K. E. (2021). 'Neuronsko strojno prevođenje', Master's thesis, University of Rijeka, Faculty of Humanities and Social Sciences; University of Rijeka, accessed 25 July 2024, https://urn.nsk.hr/urn:nbn:hr:186:436094

Lerga KE. Neuronsko strojno prevođenje [Master's thesis]. Rijeka: University of Rijeka, Faculty of Humanities and Social Sciences; Rijeka: University of Rijeka; 2021 [cited 2024 July 25] Available at: https://urn.nsk.hr/urn:nbn:hr:186:436094

K. E. Lerga, "Neuronsko strojno prevođenje", Master's thesis, University of Rijeka, Faculty of Humanities and Social Sciences, Rijeka; University of Rijeka, Rijeka, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:186:436094

Metadata
TitleNeuronsko strojno prevođenje
Title (english)Neural Machine Traslation
AuthorKošuta Estera Lerga
MentorLucia Načinović Prskalo (mentor)
Committee memberVanja Slavuj (predsjednik povjerenstva)
Committee memberLucia Načinović Prskalo (član povjerenstva)
Committee memberMarija Brkić Bakarić (član povjerenstva)
GranterUniversity of Rijeka
Faculty of Humanities and Social Sciences
(Department of Italian Language and Literature)
Rijeka
GranterUniversity of Rijeka
(Faculty of Informatics and Digital Technologies)
Rijeka
Defense date and country2021-07-19, Croatia
Scientific / art field,
discipline and subdiscipline		SOCIAL SCIENCES
Information and Communication Sciences
Scientific / art field,
discipline and subdiscipline		HUMANISTIC SCIENCES
Philology
Abstract
Nedavna otkrića u području umjetne inteligencije utjecala su na sva zanimanja, a donijela su izrazite promjene u radu prevoditelja. Googleova AutoML Translate usluga koja omogućava korisnicima da treniraju Googleove sustave temeljene na neuronskom strojnom prevođenju (GNMT) koristeći svoje vlastite prevoditeljske memorije, smatra se jednim od najznačajnijih novih proizvoda pokrenutih 2018. godine u području neuronskog strojnog prevođenja. Osim analiziranja najboljih načina i alata za stvaranje i oblikovanje prevoditeljskih memorija, rad se bavi analizom rada AutoML Translate platforme na primjeru pet različito pripremljenih prevoditeljskih memorija. Prikazani su rezultati dobiveni u navedenim slučajevima od kojih su većina dosegli izvrstan uspjeh koji je prema Googleovoj bilingvističkoj evaluaciji opisan kao model koji proizvodi visoko kvalitetan neuronski strojni prijevod.
Abstract (english)
Recent discoveries in the field of artificial intelligence have impacted various professions, including translation, and have resulted in significant changes in the work of translators. One of the most significant new products announced in 2018 is Google's AutoML Translate service, which allows users to train Google's systems based on neural machine translation (GNMT) with their own translation memories. The paper not only examines the best methods and tools for producing and structuring translation memories, but it also examines the AutoML Translate platform using five different translation memories. The preceding examples' findings are reported, with the majority of them great success, which Google's bilingual review describes as models that create high-quality neural machine translation.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:186:436094
Study programmeTitle: Italian Language and Literature (double major)
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra talijanskog jezika i književnosti (magistar/magistra talijanskog jezika i književnosti)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Public noteDrugi studijski program je informatika.
Created on2021-11-19 09:23:28