Bangla Document Categorization Using Deep RNN Model with Attention Mechanism

Text classification is a strategy where the document categories are decided. Report classification is the assignment of relegating archives to a predefined space category. In this paper, a new method has classified Bangla text documents which consist of two deep learning models in terms of Bangla text categorization. The main purpose of this research is to text classification from Bangla text documents. This system can easily categorize from a large number of text documents. Firstly, the collected data was processed for the Bangla text document. Secondly, by designing the model architecture training data it was fitted out into the model. Finally, by calculating the accuracy and F1-score on the testing dataset, the model performance was evaluated. The deep learning recurrent neural network attention layer and recurrent neural network with BiLSTM attained 97.72% and 86.56% accuracy, respectively, which is compared to the other well-known classification algorithms in Bangla text classification.