Sentiment and retweet analysis of user response for early fake news detection

Social media users have the freedom to propagate information, but this freedom has come with a price as social media has also become a channel to disseminate false information. It is thus vitally important to detect fake news in social media based on limited data available during the early stage of information propagation. This study introduces a novel approach to fake news detection and explores features e.g., retweet rate and sentiments based on the intuitions that fake news spreads faster and attracts more sentiment respectively to classify news as fake or real. The study contributes to fake news early detection by analyzing the feature sets associated with news posts on social media using limited data. The paper highlights the challenges of relevant feature sets to improve fake news detection accuracy.