ARCHIVES

The rapid growth of digital media and online platforms has significantly increased the spread of information, making fake news a serious challenge in today’s society. Detecting fake news manually is difficult due to the large volume of data and the speed at which misinformation spreads. To address this problem, this study proposes a deep learning-based fake news detection system using Long Short-Term Memory (LSTM) and Bidirectional Long Short-Term Memory (BiLSTM) models. The system utilizes Natural Language Processing (NLP) techniques for preprocessing textual data, followed by feature extraction using tokenization and sequence padding. The LSTM model captures sequential dependencies in text, while the BiLSTM model enhances contextual understanding by processing data in both forward and backward directions. The models are evaluated using standard performance metrics such as accuracy, precision, recall, and F1-score. Experimental results show that the LSTM model achieves an accuracy of approximately 99.17%, while the BiLSTM model achieves a slightly higher accuracy of approximately 99.19%, demonstrating improved performance due to bidirectional context learning. The results also indicate high precision, recall, and F1- score values, confirming the model’s reliability and effectiveness. Overall, the findings demonstrate that deep learning models are highly effective for fake news detection, with BiLSTM providing better classification performance compared to LSTM.