Fake news Detection on online Social Media

In today's digital age, social media platforms are key for information sharing, but they also facilitate the rapid spread of fake news, especially during events like the COVID-19 pandemic. A system has been developed to address this issue by categorizing news articles into six categories, from "true" to "pants on fire." Data is gathered from diverse sources like Facebook, Twitter, YouTube, and trusted organizations such as WHO and UNICEF. Techniques like Principal Component Analysis (PCA) and machine learning algorithms, including Bi-LSTM neural networks with attention mechanisms, help improve detection accuracy. Despite challenges with multi-class datasets, the system achieved 51% accuracy and a 44.9% F-score. The system also assesses the credibility of news sources and authors by evaluating social media activity and potential biases. Further improvements are sought to refine performance and expand across social media platforms.