Bangla News Classification Employing Deep Learning

In this paper, we have introduced a deep learning model for the classification of Bangla news articles as it is an important task for maintaining and managing Bangla news articles. The proposed model combines a hybrid Recurrent Neural Network (RNN) comprising Bi-directional Long-Short Term Memory (LSTM) and Bi-directional Gated Recurrent Unit (GRU) algorithms to improve classification accuracy. We tested the performance of proposed classifier with traditional machine learning techniques e.g. naïve Bayes and decision tree induction. The experimental results indicate that the deep learning models achieved almost 90% accuracy. This study provides valuable insights into the performance of various machine learning and deep learning techniques for Bangla news classification and highlights the potential of deep learning for this task.