Network Intrusion Detection via a Hybrid of Genetic Algorithms and Principal Component Analysis

In recent years network intrusion has been a problem of great concern to the design of secure computing systems. To detect an intrusion and the type of intrusion, it is desired to characterize an intrusion by its features so that future intrusions can be classified based on these features. In this paper, we present a novel method driven by the genetic algorithms and principal component analysis. This hybrid approach detects intrusions with an accuracy better than the best available till date, for several simulated tests conducted. It is hoped that this technique can also be used to detect the class of intrusion.