magistrsko delo
Rožle Reven (Author), Špela Vintar (Mentor)

Abstract

V magistrskem delu preučujem razlike v kognitivnih obremenitvah pri prevajalcih med prevajanjem in popravljanjem strojnega prevoda. Izvedel sem preizkus, ki je vključeval prevod in popravo strojnega prevoda dveh kratkih besedil iz angleščine v slovenščino, ki so ga opravili različno izkušeni prevajalci. Izbrani dve besedili v angleščini sta bili prevedeni s pomočjo nevronskega strojnega prevajalnika eTranslation, uradnega strojnega prevajalnika Evropske unije. Informacije, potrebne za določanje kognitivnega napora, so bile pridobljene s pomočjo vprašalnikov in Inputloga, programa za beleženje tipkanja. Namen tega dela je ugotoviti natančno razliko v kognitivnih procesih med prevajanjem in popravljanjem strojnega prevoda z empiričnimi podatki, ki so podkrepljeni s subjektivnimi odgovori iz vprašalnikov. Ključni del je skupna analiza odgovorov iz vprašalnikov in podatkov iz Inputloga, ki razkriva, da vsi bistveni parametri dokazujejo višjo stopnjo kognitivnega napora pri prevajanju od začetka v primerjavi s popravljanjem strojnega prevoda.

Keywords

strojno prevajanje;nevronski strojni prevajalnik;kognitivni napor;Inputlog;nevronsko prevajanje;magistrska dela;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FF - Faculty of Arts
Publisher: [R. Reven]
UDC: 81'25:81'322.4(043.2)
COBISS: 83545347 Link will open in a new window
Views: 186
Downloads: 28
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Analysis of Cognitive Effort in Translation and Post-Editing of Machine Translation
Secondary abstract: The master's thesis studies the differences of the translator's cognitive effort during translation and the post-editing of machine translation (PEMT). The experiment conducted involved the translation and PEMT of two short texts from English into Slovene by translators with various levels of experience. The selected texts were automatically translated with the neural machine translator (NMT) eTranslation, the official machine translator of the European Union. In order to determine cognitive effort, the information was obtained using questionnaires and a keystroke logging program – Inputlog. The aim of this work is to establish the exact difference in cognitive processes between translating and PEMT using empirical data that is supported by subjective answers from questionnaires. The key part of the thesis represents the analysis of the collected data that reveals that all essential parameters demonstrate a higher level of cognitive effort in translation from scratch compared to PEMT.
Secondary keywords: machine translation;neural machine translation (NMT);cognitive effort;post-editing;Inputlog;neural translation;master's theses;
Type (COBISS): Master's thesis/paper
Study programme: 0
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univerza v Ljubljani, Filozofska fak., Odd. za prevajalstvo
Pages: 68 f.
ID: 13781575