Google vs. Instagram Machine Translation: Multilingual Application Program Interface Errors in Translating Procedure Text Genre
Anisa Larassati, Nina Setyaningsih, Raden Arief Nugroho, Valentina Widya Suryaningtyas, Setyo Prasiyanto Cahyono, Stephani Diah Pamelasari · 2019 International Seminar on Application for Technology of Information and Communication (iSemantic) · 2019
The field of translation has benefited from the advance of technology and the Internet nowadays. Hence, it is common to find online machine translation tools in the cyber world. Google Translate and Instagram translation feature, a multilingual application program interface (API), are cases in point. This research focuses on the errors in how procedural texts in Instagram posts are translated into Indonesian language by using Google Translate and API. The translation error typology presented by the American Translators Association (ATA) was used as a framework of analysis. The result shows that Instagram higher errors were made by Instagram than Google Translate, with Terminology error (T), Syntax error (SYN), and Literalness (L) error being the most notable error categories in both translation machines. There is an interrelation among these three. Terminology errors were caused by the Literalness of the text elements, creating the translation of terms not suitable for the context. This research suggests that with improvements, Google Translate has a possibility to be one of the most reliable free MTs in the future. Also, the field of linguistics may become a fruitful contribution in the betterment of machine translation results.