TF-IDF Based Fake News Classification: A Lightweight Machine Learning Study with Linear SVM Dominance

Because of the prevalence of fake news in the current era that is spread in online media, this is causing fake news detection to be a huge challenge. This is because there is an abundance of fake news on social media channels, which misleads readers and provides them with fake information. Automatic fake news detection using machine learning algorithms has hence become an important research topic. In this research, we propose a model to detect fake news automatically. We make use of the Kaggle datasets that contain news data from both genuine sources and fake news websites. We perform TF-IDF vectorization, whereby we assign values to words based on their frequencies.