FAKE NEWS DETECTION BASED ON SUPPORT VECTOR MACHINE

Fake news or rumour can be a made up story or fabricated information created for the purpose of propaganda, political advantage or competitive edge. Content of fake news is created by individuals or organization and propagated via the social media networks which are largely underpinned by the cyberspace or internet. The cyberspace is an unregulated medium thereby, giving fake news propagation tremendous speed, an excessive large audience and convenience in tenns of cost and accessibility. The intention of the perpetrators may be to gain political advantage, commercial or economic edge or for other reasons. Although, a lot of extant fake news detection model have tried to combat this challenges but time complexity have been a limiting factor. To address this gap, this work proposes to explore the application of natural language processing and Machine learning technique to accurately detect fake news within the shortest possible time. The dataset will be cleaned using preprocessing tools then fed into the Support Vector Machine Classifier, which is intended to accurately state the label in a giving news article.