Empirical Analysis of Machine Learning algorithms in Fake News detection

Abstract

Social media is the finest venue for thinking and expressing in the modern world. And this is the best place to share information about your identity, culture, religion, and customs. It entails an immediate information interchange that covers news from every industry. These days, social media has a big impact on how we live and how society functions. Currently, social media is the best medium for expressing your thoughts. Social media has also evolved into a channel for disseminating information about nearby events. how the locals in the other place are made aware of what is going on there. People benefit from this through learning about various cultures. However, some evil people use social media to spread their lies, which affects society and our everyday lives. Furthermore, fake news spreads like a forest fire if it is not dealt with promptly. And this bogus news offends certain individuals and occasionally sparks riots in public places. We need instruments in the modern day that can confirm any news, whether it is real or fraudulent. The current work considers a variety of machine-learning techniques for detecting false news, including Random Forest (RF), Decision Tree (DT), and Support Vector Machine (SVM). The performance evaluation was then conducted using several criteria, including F-1 score, recall, accuracy, and precision. The empirical investigation shows DT has the greatest accuracy level at 100%.