Clause Level Reordering for Myanmar-English Translation System

Abstract

Machine Translation (MT), one of the main problems to handle is word reordering. Myanmar Language and English language are linguistically different language pairs. Myanmar language is a verb final language and English language is head initial language. After Myanmar to English language translation process, translated English words are needed to reorder to produce proper order of English sentence. Sentence level reordering system consists of word level, phrase level and clause level reordering processes. In order to get the correct order of English sentence, we propose Clause Level Reordering (CLR) Algorithm and Article Checking Algorithm for Myanmar to English translation system by mapping English grammar patterns. CLR algorithm can reorder the disorder words and it can also add relative words which are not contain in the translated sentence. Article missing words are added by using Article Checking Algorithm. We evaluate the effectiveness of proposed CLR algorithm with compound and complex English sentences.