구글과 챗GPT의 한-아 의료담화 번역 비교 연구

A Study on the Comparison between Google Translate and ChatGPT in Korean-Arabic Medical Discourse Translation

This study compared and reviewed the translations of 861 Korean medical discourse sentences into Arabic generated by Google Translate and ChatGPT, aiming to assess the feasibility and limitations of machine translation for communication purposes. The analysis of translation errors revealed no significant differences between the two tools regarding vocabulary and grammar. While Google Translate accurately rendered most medical terminology for practical use, ChatGPT exhibited numerous errors, making it unsuitable for effective communication. In terms of pragmatics, Google Translate produced vocabulary meaning errors at more than double the rate of ChatGPT and subject errors five times as often. The conversational AI nature of ChatGPT facilitated the translation of omitted subjects and context-dependent meanings to some extent. However, both tools struggled to accurately convey the contextual meanings of sentences and Korean specific expressions. These findings highlight the ongoing limitations of machine translation for communication purposes.