An Example-Based Machine Translation approach for Bahasa Indonesia to English: An experiment using MOSES

Abstract

Bahasa Indonesia and English have many differences in their linguistic structure. Translating sentences from one language to another is not a straight forward task for these pair of languages. Example-Based Machine Translation (EBMT) approach which introduced as a new paradigm in machine translation field is used in this initial work of developing a Bahasa Indonesia to English machine translation. The machine translation is developed by utilizing Moses system. Experiments in translating Bahasa Indonesia to English by tuning the parameters in Moses decoder have set alight about how the effects of manipulating the weight on translation model, language model, distortion (re-ordering) and word penalty on increasing the quality of the translation.