Social media monitoring for IoT vulnerabilities

Summary

The rapid development of IoT applications and their use in various fields of daily life by enabling new services, improving the existing ones and simplifying several routine tasks, has raised the need of making them as secure as possible. By increas- ing the number of such devices, the number of offered services also increased, and as a result the number of different possible threats escalated. In our days, social media is one of the basic cyber-threat intelligence sources. Using the appropriate tools and methods, important information can be extracted from them and used to identify new vulnerabilities and exploits. In this thesis we developed a social media monitoring system that allows users to identify vulnerabilities and recent/trending exploits on IoT devices. The monitoring scheme focuses on real-time event detec- tion from Twitter streams using state-of-the-art tools from the data science domain; we rely on a combination of a tweet monitoring and machine learning classification components trained with real-world data from the Twitter stream. The proposed classifier has been extensively evaluated against competitors using real tweets and shows promising results in recognizing tweets related to IoT vulnerabilities.