English to Bangla Machine Translation Using Recurrent Neural Network

TL;DR

This paper presents an encoder-decoder recurrent neural network architecture for English to Bangla machine translation, demonstrating improved performance over previous systems through specific activation functions and model configurations.

Contribution

It introduces a novel RNN-based translation system for English to Bangla, utilizing knowledge-based context vectors and optimized activation functions, outperforming prior models.

Findings

GRU outperforms LSTM in this architecture

Best performance achieved with linear activation in encoder and tanh in decoder

Model surpasses previous state-of-the-art in cross-entropy loss

Abstract

The applications of recurrent neural networks in machine translation are increasing in natural language processing. Besides other languages, Bangla language contains a large amount of vocabulary. Improvement of English to Bangla machine translation would be a significant contribution to Bangla Language processing. This paper describes an architecture of English to Bangla machine translation system. The system has been implemented with the encoder-decoder recurrent neural network. The model uses a knowledge-based context vector for the mapping of English and Bangla words. Performances of the model based on activation functions are measured here. The best performance is achieved for the linear activation function in encoder layer and the tanh activation function in decoder layer. From the execution of GRU and LSTM layer, GRU performed better than LSTM. The attention layers are enacted with softmax and sigmoid activation function. The approach of the model outperforms the previous state-of-the-art systems in terms of cross-entropy loss metrics. The reader can easily find out the structure of the machine translation of English to Bangla and the efficient activation functions from the paper.