Fact Check Using Multinomial Naive Bayes

Pradhan, M. A.; Shinde, A.; Dhiman, R.; Ghorpade, S.; Jawale, S.

An easy access to social media platforms has made information available effortlessly and thus has increased the intricacies to distinguish between true and falsified information. The credibility or reliability on social media platforms is also at stake. It is of utmost necessary to address this as a severe issue and act on it promptly. The extensive spread of counterfeit news has the potential for creating negative impacts on vast audience. Therefore, fake news detection on social media has become a very critical agenda in today’s world. This paper proposes a prototype to detect whether a news is fake or real using the multinomial Naive Bayes algorithm and its various architectures. Furthermore, the proposed prototype is capable of handling the unstructured data as the news can be in the form of images. In addition to this, the use of Django which is a high-level Python framework that allows the development of UI very easily with multiple designing options. As there was a high need of a 24/7 working server, the system has been deployed on Amazon Web Services EC2 Server as it gave less downtime and is highly reliable. Experimentation was done on the synthetic COVID news dataset created by collecting COVID news on social platforms. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.