Comparison Between SVM and Back Propagation Neural Network in Building IDS

Recently, applying the novel data mining techniques for anomaly detection-an element in Intrusion Detection System has received much research alternation. Support Vector Machine (SVM) and Back Propagation Neural (BPN) network has been applied successfully in many areas with excellent generalization results, such as rule extraction, classification and evaluation. In this paper, we use an approach that is entropy based analysis method to characterize some common types of attack like scanning attack. A model based on SVM with Gaussian RBF kernel is also proposed here for building anomaly detection system. BPN network is considered one of the simplest and most general methods used for supervised training of multilayered neural network. The comparative results show that with attack scenarios that we create and through the differences between the performance measures, we found that SVM gives higher precision and lower error rate than BPN method.