Cross-SEAN: A Cross-Stitch Semi-Supervised Neural Attention Model for COVID-19 Fake News Detection

摘要: As the COVID-19 pandemic sweeps across world, it has been accompanied by a tsunami of fake news and misinformation on social media. At time when reliable information is vital for public health safety, related spreading even faster than facts. During times such as pandemic, can not only cause intellectual confusion but also place lives people at risk. This calls an immediate need to contain spread We introduce CTF, first Twitter dataset with labeled genuine tweets. Additionally, we propose Cross-SEAN, cross-stitch based semi-supervised end-to-end neural attention model, which leverages large amount unlabelled data. Cross-SEAN partially generalises emerging learns from relevant external knowledge. compare seven state-of-the-art detection methods. observe that achieves $0.95$ F1 Score outperforming best baseline $9\%$. develop Chrome-SEAN, chrome extension real-time