مقارنة نتائج الترجمة من اللغة الإندونيسية إلى اللغة العربية بين جوجل الترجمة و مكروسوف الترجمة و ديبل الترجمة

Abstract

This study aims to compare the results of translation from Indonesian to Arabic using three translation applications: Google Translate, Microsoft Translator, and DeepL Translator. This research is important to understand the performance of translation applications in handling languages with different structures and cultures. This study uses a qualitative descriptive method. The research data is in the form of an abstract of student research at the State University of Makassar. The analysis was carried out by comparing the translation results of the three applications based on several criteria, including meaning suitability, grammatical accuracy, and vocabulary richness. The results of the study showed that 26 errors were found in Google Translate, Microsoft Translator showed 24 errors, and in deepl translator the translator had made 24 errors. When comparing Google Translate, Microsoft, and deepl errors in three criteria, it can be concluded as follows: in the conformity of the meaning of Google Translate: 18 errors, Microsoft Translate: 17 errors, and deepl translator: 17 errors. In grammatical accuracy, Google Translate: 13 errors, Microsoft Translate: 11 errors, and Deepl Translator: 11 errors. And in the rich vocabulary of Google Translate: 10 errors, Microsoft Translator: 13 errors and deepl Translator: 13 errors. Errors in the conformity of meaning are most common in all translators, while errors in grammatical accuracy and vocabulary richness vary between them. This suggests that there are common challenges in translation applications related to understanding meaning and context, as well as grammatical accuracy and vocabulary selection.