Izgradnja sustava za neuralno strojno prevođenje

master's thesis

• Title (english): Development of a neural machine translation system
• Author: Jelena Saratlija
• Mentor: Ivan Dunđer (mentor)
• Mentor: Sanja Seljan (sumentor)
• Committee member: Vjera Lopina (predsjednik povjerenstva)
• Committee member: Sanja Seljan (član povjerenstva)
• Committee member: Ivan Dunđer (član povjerenstva)
• Granter: University of Zagreb; Faculty of Humanities and Social Sciences; (Department of information and Communication sciences); Zagreb
• Defense date and country: 2020-09-18, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Information and Communication Sciences; Information Systems and Information Science

Abstract

Konstantni razvoj tehnologije omogućava napredak u svim područjima njezine primjene. Jedno od tih područja je i strojno prevođenje čija je važnost neizmjerna u vremenima globalizacije. Potrebno je osigurati neometan protok informacija koje će svima biti jasne i razumljive. U svrhu postizanja tog cilja važno je raditi na optimiziranju sustava za strojno prevođenje. U radu će biti prikazani teorijski i praktični aspekti izgradnje sustava za neuralno strojno prevođenje te će se usporediti statistički i neuralni model strojnog prevođenja. Sustav će biti izgrađen u odabranoj domeni nad pripremljenim skupom podataka u kojoj će se zatim testirati i evaluirati.

Abstract (english)

A continuous technological change improves every aspect of our lives where it is being implemented. One of those aspects is also machine learning whose importance is immense in times of globalization. Nowadays it is necessary to ensure a consistent flow of information that is understandable and comprehensible for everyone. In order to achieve that goal, it's important to work on optimizing machine translation systems and implementing the newest technologies. This thesis is intended to represent theoretical and practical aspects of development of a neural machine traslation system and to compare statistical and neural model of machine translation. A machine translation system will be built in a chosen domain with a prepared set of data in which it will be also tested and evaluated.

• Keywords: strojno prevođenje; neuralne mreže; duboko učenje; neuralno strojno prevođenje; umjetna inteligencija
• Keywords (english): machine translation; neural networks; deep learning; neural machine translation; artificial intelligence
• Language: croatian
• URN:NBN: urn:nbn:hr:131:440947
• Study programme: Title: Information Sciences; specializations in: Computer Science (Teaching, Research), Informatology, Information Sciences (teaching), Information Sciences (research); Course: Information Sciences (research); Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra edukacije informacijskih znanosti (magistar/magistra edukacije informacijskih znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2020-09-19 09:34:09