Fake News Detection on Social Media Using Regional Convolutional Neural Network Algorithm

Abstract

As social media and content-sharing platforms have evolved; misinformation and fake news have spread like wildfire, leading people to believe harmful misinformation. In this way, they can influence public opinion, spread fear, and drive people insane. Fake news identification is a current area of research aimed at determining whether content is genuine. In addition, this has significantly increased the daily amount of information on the Internet. Information can go viral in a matter of seconds thanks to social media, which has changed the way we share and process news. Everyone now relies on many online news sources because the internet is so widely used. News quickly disseminated across millions of users in a very short period of time along with the increase in the use of social media platforms like Facebook, Twitter, etc. The spread of fake news has far-reaching effects, including the formation of skewed beliefs and the manipulation of election results in favour of particular politicians. Moreover, spammers utilise alluring news headlines as click-bait for their adverts in order to make money. To provide more accurate predictions, RCNN models are trained to identify language-driven features according to content properties. This model addresses this using an efficient feature selection method.