3rd Asia Pacific International Conference on Industrial Engineering and Operations Management

Analysis of Japanese-Indonesian Translation Using Machine Translation

Yuwan Prananta Mulyadi & Elisa Carolina Marion
Publisher: IEOM Society International
0 Paper Citations
1 Views
1 Downloads
Track: Machine Learning
Abstract

Translating Japanese to Indonesian using the Google translation engine is an easy thing. However, the results of the google translate translation must go along as possible with reasonable accuracy. This study aims to determine the accuracy of Google Translate translation in translating Japanese articles into Indonesian. The research uses combined research methods through literature studies from various literature sources. The theories used in this study are Machali's assessment theory (2009), Hartono's translation theory (2017), Bassnett-McGuire's addition and subtraction theory (1991), and the BLEU algorithm (2002). The data is taken from a speech of the Ambassador of Japan to Indonesia, taken from an official article on the embassy's website. The research found the correct semantic assessment value, incorrect semantic assessment value, type of semantic error, and the value of the BLEU algorithm assessment. Google translate can provide a correct result in terms of semantics or meaning. The average correct semantic value indicates this is around 88.03%. The assessment of the BLEU algorithm for google translate text gets an assessment result of 37.11. This average value includes a low value, which shows a significant difference in structure between the google translate text and the translator text. In other words, the results of google translate are suitable for conveying the meaning of the results of the translation. However, it is essential to improve the sentence structure of the results of Google Translations. 

Published in: 3rd Asia Pacific International Conference on Industrial Engineering and Operations Management, Johor Bahru, Malaysia

Publisher: IEOM Society International
Date of Conference: September 13-15, 2022

ISBN: 978-1-7923-9162-0
ISSN/E-ISSN: 2169-8767