Google Translate Accuracy in Translating Specialized Language From English to Bahasa Indonesia:

Abstract

This research aims to discover the accuracy level of Google Translate in English -Bahasa Indonesia translation, especially in translating specialized language in mechanical engineering terms. This study belongs to descriptive qualitative research. The primary source of the data was the output translation texts of Google Translate. The sample texts were sentences taken from the book of the mechanical engineering field. The main concerns were on word and phrase levels. Since there were only two categorizations in accuracy research, the researchers divided the number into two. A text was said to be accurate if the accurate meanings were more than 50%. Conversely, it was inaccurate if the accurate meanings were less than 50%. The results revealed that in English -Bahasa Indonesia translation, Google Translate, especially in translating specialized language in mechanical engineering terms, was accurate for word translation, but inaccurate for phrases. This statement was based on the findings of accurate occurrence percentage in Google Translate translations, which were 70.73 % for the word level and 46.37% for the phrase level. The number for words was above 50 %, but it was the opposite for phrases. The inaccuracy was indicated by four inaccuracy indicators (omission, addition, different meaning and zero meaning). Moreover, several sentences translated by Google Translate were inaccurate. This research unveiled that Google Translate translated the sentences or paragraphs based on the sentence structure from the source text and the general meaning of the words or phrases.

SNLI Indo: A recognizing textual entailment dataset in Indonesian derived from the Stanford Natural Language Inference dataset

“…The Google Cloud Translation API, which was developed by Google, was utilized in this research as the automatic translation tool because it has been proven to be adequately accurate in intelligent machine translation. The Google Cloud Translation API remains a reliable tool for translating simple sentences using standard and coherent phrase levels [10 , 11] .…”

Accuracy of Nursing Term Translation from Indonesian to English by Nursing Students

“…Language for special purposes or ESP can address at least two perspectives: the first is called didactic, because ESP is the field of language teaching. The second is that a special linguistic code is needed to describe the concepts that govern the physical world (Farahsani et al, 2021a). Meanwhile, according to (Amna & Idriani, 2019) English for special purposes is given as special required English language ability.…”

“…Besides that, Newmark also explained that there are several things related to translation, they are: the meaning of the text which contains the contents of the text in the source language, the intention of the translator includes the purpose of the translation, the target reader and the background of the text focusing on politeness and diction strategies, and the quality of the translation regarding the intent of the text in the source language (Farahsani et al, 2021a). Therefore, translation activities are not only translating words and phrases, but it is also more like the essence of the text being translated.…”

An Analysis of Translations Techniques by Using (GT) Google Translate’s English into Indonesia in English Simple Sentences

“…GT uses human translations to identify the best possible translation pattern for a given text by scanning through multiple sources (Farahsani, Rini, & Jaya, 2020). A Google Translation's degree of accuracy is directly proportional to the number of human-translated texts that the service searches for.…”