Genetic algorithm combined with support vector machine for building an intrusion detection system

Abstract

In this paper, we develop an intrusion detection system (IDS) based on machine learning. We employ genetic algorithm (GA) along with Support Vector Machine (SVM) for automatically determining the appropriate set of features. The idea is then developed into a fully functional IDS. Experiments of testing the IDS on the benchmark KDD CUP 99 datasets are presented. Results show encouraging performance that opens a avenue for further research.