Institutional Repository of Vilnius University
Comparative analysis of machine and artificial intelligence translations of poetic metaphors /
Abstract
dc:descriptionThis study focuses on translation quality evaluation of poetic metaphors, using ChatGPT 4.0 and DeepL online translator. Chosen poetry was translated from English to Lithuanian, and poetic metaphors were identified afterwards. Using the Multidimensional Quality Metrics criteria, both versions of the translations were evaluated separately. Main mistakes with differences were identified. This thesis analyses the specific differences and examples of both translation tools. In this paper it is discussed how different error types affect the poetic metaphor translation and analyses the shortcomings and advantages of the tools. The aim of the thesis was achieved by collecting data on the error number in each category and analysing specific examples. The results of the analysis revealed that both tools have their shortcomings, especially in selecting the appropriate style and terminology. However, ChatGPT overall outperformed DeepL online translator. The most significant difference was found in the categories of accuracy and style, whereas in terminology, both of the tools performed the same. In terms of grammar, though ChatGPT performed slightly better, both of them made minimal mistakes. Therefore, even though the tools might not be accurate when translating figurative or poetic speech, they have demonstrated accuracy in terms of semantics and grammar.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Dambrauskaitė, Liucija,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD239014549&prefLang=en_US