Chapter 8 - Machine Learning Techniques for Threat Modeling and Detection

https://doi.org/10.1016/B978-0-12-811373-8.00008-2Get rights and content

Abstract

The major goal of this chapter is to overview and present the practical realizations of the bio-inspired concepts for cyber security. Hereby, we do not discuss the bio-inspired concepts for cyber security on a high and abstract level, but we present our own concrete solutions and practical implementations and also briefly mention other relevant works. Our goal is to prove that the bio-inspired techniques can be implemented to cyber security and that readiness level of such technology is constantly increasing. Particularly, we have investigated and presented the practical solutions for the evolutionary-based optimization techniques, collective intelligence, and techniques that mimic social behavior of species. In this chapter, we present and focus on our own results and give references to our past and ongoing cyber security projects where we successfully implemented different nature-inspired solutions. The proposed genetic algorithms improve detection of SQL injection attacks and anomalies within HTTP requests. Similarly, the proposed ensemble of classifiers and correlation techniques allow for the improved cyberattack detection. Furthermore, the collective intelligence concept has been successfully implemented in the Federated Networks Protection System.

References (0)

Cited by (0)

View full text