Asaf Shabtai

Fake news is a long-lasting problem which has drawn significant attention in recent years. There is a growing need for tools and methods to control the spread of misinformation through online social media. Machine learning methods have been utilized to pinpoint linguistic patterns, influential accounts, or spreading dynamics associated with misinformation. In this paper, we present an automated process for training fake news classifiers based on multiple families of features extracted from social media. In addition to the high accuracy of the trained machine learning classifiers, our results show that online social media users are aware of deceptive content and can often provide reliable feedback for the detection of fake news.