Deep Learning-based Text News Classification using Bi-directional LSTM Model

The advancement of technology has resulted in increased news consumption through various online platforms and social media channels. Hence, there is a requirement for intelligent systems that may categorize the news that exists on the social media/online platform. Hence, this paper explores creating a news classifier utilizing a Bidirectional Long ShortTerm Memory (Bi-LSTM) deep learning model, which is a type of recurrent neural network (RNN) specifically designed to overcome the challenge of vanishing gradient often faced in deep learning. The machine learning was performed on a Kaggle dataset with over 400,000 records of news stories from various publishers. Natural language processing techniques such as stemming, one-hot encoding, vectorization, and embedding are used to prepare the data for machine learning. The multiclass classification is achieved and the performance of the model is analyzed.