A Comparative Analysis of Google Translate and ChatGPT Results Based on Translation Quality Assessment (TQA)

Abstract

The development of digital technology today has a major impact on various aspects of human life. In bridging intercultural communication, translation has a significant role that cannot be separated from the help of technology. Technology also has brought changes in the way translation is done, including helping individual translators to increase efficiency. For example, with the help of machine translation, Google Translate continues to develop transformers and AI. On the other hand, ChatGPT, built on artificial intelligence, allows translation. Both adopt NLP. This study aims to determine the quality of the translation of both based on the TQA proposed by House. In addition, it assesses the level of similarity of the translation results and errors that occur based on the translation technique used. Translation samples were selected based on purposive sampling techniques, namely from a report chapter available on the internet in English. The samples were translated into Indonesian using Google Translate (GT) and ChatGPT (CG) and then analyzed using TQA. The results show that GT and CG have high similarities and insignificant error rates. This is possible because GT and CG continue to be developed over time to respond to and translate languages that are increasingly natural and resemble human language.

Ähnliche Arbeiten