Fake News Detection Using Machine Learning: A Comparative Analysis of Algorithms

Abstract

With the rapid proliferation of fake news, distinguishing between real and fabricated information has become a critical challenge in the digital age. Fake news not only manipulates public opinion but also poses a significant threat to social stability. This study explores the application of machine learning techniques for the automatic detection of fake news articles. A labelled dataset comprising both genuine and fake news is utilized, incorporating preprocessing techniques such as data cleaning, text normalization, and feature extraction. The research implements and evaluates multiple machine learning models, including Logistic Regression, Naïve Bayes, and Support Vector Machines (SVM), comparing their performance based on standard evaluation metrics such as accuracy, precision, recall, and F1-score. The experimental results demonstrate that machine learning models significantly enhance fake news detection, with DistilBERT achieving a state-of-the-art accuracy of 99.90%, outperforming traditional approaches. The study underscores the effectiveness of Natural Language Processing (NLP)-based techniques in improving classification accuracy. The findings contribute to the growing body of literature on automated misinformation detection and highlight potential avenues for further research, such as integrating metadata-based features, ensemble learning approaches, and real-time detection systems. These insights serve as a foundation for developing more robust and scalable solutions for combating misinformation in digital media.