Aristotle University Of Thessaloniki (AUTH)
Αυτόματη μετάφραση μεταφορών και ιδιωτισμών από τα ιταλικά στα ελληνικά
Abstract
dc:descriptionIn the era of multilingualism and globalization, the demand for translation services is continuously increasing. Translation technology, including machine translation, significantly contributes not only to translate faster but also provide better quality translations. The big demand for translation services creates the need for machine translation post-editors to ensure that the translations transmit a fully understandable message in the target language. The term "machine translation post-editing", which should not be confused with translation, refers to the process of optimizing the quality of a machine-translated text. It’s a new role that a translator acquires through training, as it requires special handling and better memory management since the machine translation outputs contain different types of errors than the translation memories. This doctoral thesis falls within the field of Applied translation studies which is subdivided into translator training, translation aids and translation criticism. The purpose of our study is to investigate, through translation evaluation, whether translation technology and, in particular, the use of translation aids, machine translation and concordance search improve the quality of translations and if it is a useful aid for the translator's work. Through the literature review, we describe the advantages and disadvantages of translation technology and its contribution to the translation process. We examine the modern translator’s profile and we refer to the differences between machine translation post-editing and translation in order to propose a translation strategy methodology that will help translators produce effective communication translations, when post-editing machine-translated raw output. To investigate the thesis cases, we designed an experimental setup in which five translation students were involved. We created a questionnaire with general questions and translation tasks, using authentic Italian texts which include idioms and metaphors with the multi-meaning verb fare (= to do). The participants were asked to fill in the questionnaire and translate the tasks into Greek emphasizing on the translation of the idioms and metaphors. Part of them, i.e. 10 sentences, were translated in a conventional way, therefore, without the support of any translation aid. For the remaining 20 sentences, translators were provided with machine-translated suggestions as well as results of concordance search for each idiom. From the qualitative and quantitative analysis of the translations we conclude that the machine-translation raw output after the post-editing stage is better in quality compared to the translations for which the translators had no means of aid. Although the results are encouraging, there are cases of translations that are stylistically, pragmatically and/or morphologically lagged, resulting in loss of meaning in the target language. We are also involved in the study of translators' performance using translation technology, translation unit management, working memory management as well as evaluation of translation quality through the translation declaration of the subjects.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzitheodorou, Konstantinos
- Χατζηθεοδώρου, Κωνσταντίνος
Subjects
dc:subject × 28- Επιμέλεια αυτόματης μετάφρασης
- Μεταφραστικά βοηθήματα/ Εργαλεία μετάφρασης
- Εκπαίδευση μεταφραστών
- Κριτική μετάφρασης
- Επεξεργασία φυσικών γλωσσών
- Επιχώρια προσαρμογή
- Machine translation post-editing
- Translation aids
- Translator training
- Translation criticism
- Natural language processing
- Localization
- Post-editing di traduzione automatica
- Sussidi alla traduzione
- Formazione del traduttore
- Critica della traduzione
- Elaborazione del linguaggio naturale
- Localizzazione
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Ανθρωπιστικές Επιστήμες και Τέχνες
- Γλώσσα και Λογοτεχνία
- Άλλες Ανθρωπιστικές Επιστήμες
- Natural Sciences
- Computer and Information Sciences
- Humanities and the Arts
- Languages and Literature
- Other Humanities
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/47250