2026-04-16T23:26:49Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4996 2021-12-13T02:41:55Z 1582963302567:1597824273898 user-ucsy

Improving English-to-Myanmar Statistical Machine Translation by using Recurrent Neural Network Language Model and Hierarchical Reordering Zin, May Myo Soe, Khin Mar In this paper, a comparison between neuralbased network that adopts Recurrent Neural Network(RNN) language model and statistical based one withN-gram language model is conducted for English-to-Myanmar phrase-based statistical machine translation(PBSMT). In this comparison, lexicalized reorderingmodels such as word-based, phrase-based andhierarchical orientation models are used as anadditional reordering model to investigate the overallperformance of PBSMT. The perplexity valueevaluation of both language models showed that theuse of RNN obtains a more excellent result. Accordingto the obtained BLEU and RIBES scores andadditional human visual inspection, the English-to-Myanmar PBSMT with RNN language model andhierarchical reordering model is the best one in termsof improving adequacy and fluency. 2017-02-16 http://hdl.handle.net/20.500.12678/0000004996 https://meral.edu.mm/records/4996