Enhanced Fake News Detection with the Aid of Improved Spider Monkey Optimization-Based Optimal Feature Selection and Deep Neural Network

Abstract

Fake news has become a significant problem in recent years, leading to widespread misinformation and public manipulation. This research focuses on developing an effective fake news detection model using advanced machine learning and deep learning techniques. Existing methodologies face challenges such as poor performance with large datasets, noise, and limited generalization. The proposed solution integrates pre-processing, feature extraction, optimal feature selection via Spider Monkey Optimization (SMO), and a deep neural network (OAF-DNN) with optimized activation functions. The model's performance will be validated using publicly available datasets and analyzed through various evaluation metrics. This study aims to enhance the accuracy, precision, and detection of fake news.