A Comparative Study of Hybrid Machine Learning Approaches for Fake News Detection that Combine Multi-Stage Ensemble Learning and NLP-based Framework

Fake News has been spreading widely throughout the world as the booming internet era has started worldwide. Now, more people have access to the Internet than ever, which has led to a significant rise in spreading fake news. So, to solve this issue, it would be highly impossible to manually remove every phony news article. To tackle the above problem of checking on information related to the source, content, or news publisher to categorize it as genuine or fake, we take the help of Machine Learning to classify the information on the web as True or False. Therefore this paper explores the different types of ML classifiers to detect fake news. Therefore, this study will use textual properties of the news dataset we took from Kaggle to distinguish a piece of news as fake or real. Furthermore, with these properties, we will train our model using different ML classification algorithms to evaluate the performance of the dataset collected.