Machine Translation Models and its Challenges when Applied on Various Domain Corpus

Abstract

Machine translation is an art of language engineering and is a sub branch of computational linguistics that carries out automated translation of human languages from one source language to another target language. There are many approaches to machine translation beginning from the basic existing statistical MT and rule based  MT to the present neural machine translation approaches. Even though it is proved that SMT has noticeable less performance and accuracy than neural machine translation, it is also proved that SMT is still more suitable for different databases of text. So this paper would give a systematic review on various research done in machine translation before to neural machine translation.