The accuracy of machine translation is still debatable. This study aims to compare the accuracy of Google Translate and BARD in translating scientific journal abstracts. The main focus is to evaluate the extent to which these two platforms can preserve the meaning and quality of translation in the abstracts of Adabiyyāt journal. This research utilizes two main theories in evaluating translation accuracy, namely the Human-mediated Translation Edit Rate (HTER) by Snover (2006) to identify translation errors and the Translation Quality Index (TQI) by Schiaffino & Zearo (2005) to calculate overall accuracy results. The results of this study show that GT is proven to be more accurate than Bard in translating abstracts from scientific journals. This can be seen from the final result of GT's calculation, which is greater than that of Bard. GT scored 99,7% in total, while Bard only got 99,1%.
Copyrights © 2024