Are They There Yet? Revisiting the Accuracy Level of Computer-Generated Translation
A Suparman (1*), M T Hidayat (1), C U Abdullah (2), Eri Subekti (3), H Apriliani (1)
1) Institut Pendidikan Indonesia, Jl. Terusan Pahlawan No. 32, Garut, Jawa Barat 44151, Indonesia
*asep.suparman[at]institutpendidikan.ac.id
2) Universitas Pendidikan Indonesia, Jl. DR. Setiabudhi No. 229, Bandung, Jawa Barat 40154, Indonesia
3) Universitas Langlangbuana, Jl. Karapitan No. 116, Bandung, Jawa Barat 40261, Indonesia
Abstract
Speed, cost, and quality have been the central issues in the translation industry. While computers can offer fast and economical translation, their ability to produce accurate translation at a publishable level like human translators may do is still questionable. In the present study, we evaluate the accuracy of English to Indonesian translation generated by Google Translate. Freely accessible English texts of various types were chosen to heuristically exemplify the units of analysis. Our findings lead to a conclusion that while Google Neural Machine Translation system is quite an improvement to the conventional phrase-based translation system, highly accurate translation does not seem to be achieved by Google Translate in the near future.
Keywords: translation accuracy, machine translation, Google Translate, English to Indonesian
Topic: Computer Science