This paper presents the NICT's supervised and unsupervised machine translation systems for the WAT2019 Myanmar-English and Khmer-English translation tasks. For all the translation directions, we built state-of-the-art supervised neural (NMT) and statistical (SMT) machine translation systems, using monolingual data cleaned and normalized. Our combination of NMT and SMT performed among the best systems for the four translation directions. We also investigated the feasibility of unsupervised machine translation for lowresource and distant language pairs and confirmed observations of previous work showing that unsupervised MT is still largely unable to deal with them.
This paper proposed a unified approach for Myanmar Word analysis using Finite State Automata (FSA), Rule Based Heuristic Approach and Statistical Approach. Myanmar has no inter-word space and it make the tokenizing task difficulties. Therefore, to recognize the word, we implement with FSA. Segmentation is a major problem because of no delimiter. If there were errors in segmentation, this will cause subsequence failure in further NLP processes. Segmentation is also an essential preprocessing task for Natural Language Processing, such as Machine Translation, Information Retrieval etc. In this system, the Rule Based Heuristic Approach and Statistical Approach are used with corpus based dictionary. Evaluation results showed that the method is very effective for the Myanmar language.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.