Detecting Fake News using Machine Learning Techniques

Particularly in the digital era, the speed and scope of false news distribution create a major security threat that calls for the creation of efficient detection systems to limit it. The performance of two machine learning models in the identification of false news: K Nearest Neighbours and Naive Bayes, is compared in this work. The goal of this work is to improve the false news detection accuracy. For every model, accuracy, precision, recall, and F1-score were obtained in assessing model performance. With 96.44% instead of KNN's accuracy of 93.75%, Naive Bayes showed superior accuracy. The possible use of the Naive Bayes method to get better prediction accuracy for the identification of false news is investigated in this work. This study offers a strong and scalable detection system to try to counteract false news.