Computational Challenges and Recent Advancements in Automated Fake News Detection

"Fake news" is now viewed as one of the greatest threats to democracy, freedom of expression, and journalism. Massive spread of fake news has weakened public trust in governments and its potential impact on various political processes, e.g., the "Brexit" referendum or the equally divisive 2016 U.S. presidential election, is yet to be realized. We will briefly review some of the modern computational techniques for fake news detection, along with some of the current challenges that these methods face. We will discuss some recent advancements to tackle these challenges, with particular focus on fake news early detection, multimodal fake news analysis, modeling the intent of fake news spreaders, and the lack of data.