Analyzing the Perceived Severity of Cybersecurity Threats Reported on Social Media

This paper analyzes the actual severity of a cybersecurity threat through the language that is used to describe it on social media platforms such as Facebook or Twitter. The keywords “DDoS” and “vulnerability” were tracked from December 2017 to July 2018, and were annotated by the researchers to determine the actual severity of the threat. The paper shows that the perceived severity of threats on social media posts can indeed be used as an indicator of the presence of software vulnerabilities. Furthermore, the presence of these posts can be used to predict the presence of real-world exploits, as well. The paper also shows that automatic classifiers can be used to aggregate these posts to help gauge the severity of a threat.