The Influence of Word Vectorization for Kawi Language to Indonesian Language Neural Machine Translation

People relatively use machine translation to learn any textual knowledge beyond their native language. There is already robust such as Google translate. However, the language list has only covered high resource English, France, etc., but not for Kawi Language one of local languages used in Bali's old works literature. Therefore, it necessary study development from more active user Indonesian make easier learning access young learner. The research developed neural (NMT) using recurrent network (RNN) based models and analyzed influence word vectorization Word2Vec performance on BLEU scores. result shows that indeed significantly increases NMT performance, Long-Short Term Memory (LSTM) with attention mechanism highest scores equal 20.86. still could achieve par those human experts translation. On other hand, this initial be reference future NMT.