Intrusions detection based on Support Vector Machine optimized with swarm intelligence

Intrusion Detection Systems(IDS) have become a necessary component of almost every security infrastructure. Recently, Support Vector Machines (SVM) has been employed to provide potential solutions for IDS. With its many variants for classification SVM is a state-of-the-art machine learning algorithm. However, the performance of SVM depends on selection of the appropriate parameters. In this paper we propose an IDS model based on Information Gain for feature selection combined with the SVM classifier. The parameters for SVM will be selected by a swarm intelligence algorithm (Particle Swarm Optimization or Artificial Bee Colony). We use the NSL-KDD data set and show that our model can achieve higher detection rate and lower false alarm rate than regular SVM.