Bad Actors in Social Media

Abstract

Bad actors seriously compromise social media every day by threatening the safety of the users and the integrity of the content. This keynote speech will give an overview of the state of the art social network analysis, data mining, and machine learning techniques to detect bad actors in social media. More specifically, we will describe both general methods that are platform independent or valid for any malicious user, and methods specific to a particular social media (Twitter, Facebook, Slashdot, Wikipedia, Instagram) and/or a given type of bad actor (bots, spammers, trolls, vandals, cyberbullies). We will group these methods into four broad categories, namely (i) active methods, (ii) content-based, (iii) social network-based, and (iv) behavior-based, and show their effectiveness in enforcing cybersafety.