Neural approach-based quality estimation in improving translation of English to Hindi using machine translation under data science

Quality estimation could be a method for automatically indicating the standard of computational linguistics results without looking forward to human reference translations, i.e. determining how a machine translation system can produce good or bad translations without need human intervention. The main aim of this paper is to demonstrate the concept used to deal with the problem of estimating on a Hindi and English combined language pairs. To perform the proposed technique, we are using DL based quality estimation feature extraction and therefore, in this paper, we perform experiments on a set of multiple features along with the various Neural RNN models. The first part of implementing the recurrent neural networks generates the quality information about whether each word in translation is properly translated. The second part is that using the accuracy and loss of all the RNN models predicted is used to compare the quality of translation of each of the models. We apply these models on the English to Hindi taken from manythings.org website.