FAKE NEWS DETECTION USING CONVOLUTION NEURAL NETWORK ON SOCIAL PLATFORMS

Abstract

The majority of smart phone users choose social media instead of internet when it comes to reading the news. The news published on news websites, which serve as the source of reliability. Fake news threatens logical truth because it is difficult for people to discriminate between genuine and false information, which damages journalism, democracy and the trust people have in governmental institutions. Evolving technologies have made it imperative to create approaches that can limit the dissemination of false information that could negatively impact society in any way. Online users tend to be vulnerable and will generally believe everything they encounter on social networking websites to be trustworthy. In order to maintain robust internet media and informal organisations, automating counterfeit news identification is fundamental. To take rumours seriously and present them as news is detrimental to society. Stopping rumours is urgently needed, particularly in developing countries. Instead, attention should be paid to accurate, reliable news reports. We provide a model for recognising forged news that is a computational stylistic study based on NLP, which can be used to efficiently deploy deep learning algorithms like ANN and CNN algorithm to identify fake stories in texts obtained from social platforms.