Fake News Recognition: A Machine Learning Approach for Text Analysis using Hyperparameter Tuning

The internet has revolutionized communication and has come an ingrained part of people’s lives, with a significant number of individualities exercising it for colorful purposes. Social media platforms have emerged as popular online destinations, allowing users to share posts and disseminate information. Unfortunately, these platforms often lack robust verification processes for users and their content. Consequently, some individuals exploit this vulnerability by spreading through these sites, bogus news. Such false information can be used as a tool for propaganda, targeting individuals, societies, organizations, or political parties. Given the overwhelming volume of online content, it becomes nearly impossible for humans to manually detect all instances of fake news. Therefore, there is a pressing need to develop classifiers for machine learning capable of automatically identifying and flagging fake news. This research study aims to explore the utilization of machine learning classifiers based on hyperparameter tuning for false news classification, shedding light on the various methodologies and techniques employed in this domain.