In the contemporary information age, most of our mornings begin with reading news and postings from various applications such as in short, Facebook, Twitter, and others, which differ from traditional printed media in that the news or posts’ validity cannot be ensured. This situation necessitates the development of automatic ways for detecting and removing fraudulent content from the internet so that people are not misled by hoaxes propagated with the goal of profiting from the situation. To detect fake news, we examine various machine learning-based techniques and evaluate overall performance against fake news propaganda. We conducted a comparative analysis of the different deep learning (DL) and machine learning (ML) based systems we developed. Our proposed system gave 99.8% accuracy with the BERT (Bidirectional Encoder Representations from Transformers) word-embedding on the ‘ISOT’ fake news dataset.