Machine Translation and Post-Editing in Special Education Research Titles: A Comparative Study of Google Translate and ChatGPT-4
This qualitative study compares machine translations and post-editing of academic titles in the field of special education and rehabilitation from Serbian to English using Google Translate and ChatGPT-4. First, thirty-eight titles from two conference proceedings were machine-translated by both tools...
Salvato in:
| Pubblicato in: | URI:https://journals.openedition.org/asp, |
|---|---|
| Autore principale: | |
| Natura: | Article ou chapitre numérique |
| Pubblicazione: |
ASp
2025
|
| Soggetti: | |
| Accesso online: | Accès Université d'Orléans et IFPM Accès Université d'Orléans et IFPM |
| Riassunto: | This qualitative study compares machine translations and post-editing of academic titles in the field of special education and rehabilitation from Serbian to English using Google Translate and ChatGPT-4. First, thirty-eight titles from two conference proceedings were machine-translated by both tools. Nineteen title pairs were then selected for the post-editing process conducted by a professional and experienced translator. Two main criteria were considered in the analysis: accuracy of specialized terminology and overall fluency and readability. The findings show that ChatGPT-4 produced more natural and fluent translations, with fewer redundancies and issues related to specialized terminology, compared to Google Translate. Still, human post-editing was necessary for both translation tools to ensure natural flow, clarity, and terminological precision. While machine translation is time-saving and cost-effective, it can only be used as a starting point in the translation process, where professional post-editing remains indispensable for highly specialized contexts. Given the small exploratory dataset, the findings should be interpreted with caution. |
|---|