Skip to Main Content
Supervised anomaly intrusion detection systems (IDSs) based on Support Vector Machines (SVMs) classification technique have attracted much more attention today. In these systems, features of instances and the characteristic of kernels have great influence on learning and predict results. However, selecting feasible features and kernel parameters can be time-consuming as the number of features and the parameters of kernel increase. In this paper, a quantum-inspired immune evolutionary algorithm (QIEA) based parameter optimization approach is introduced to solve these problems. The mixtures of kernels are used for improving the learning and predict performance of SVM. At the same time, the real-coded chaotic QIEA is used for optimizing the parameters of mixtures of kernels. The KDDCuppsila99 dataset was used for performance comparison and the experiment results show that the proposed method is efficient competent with the Differential Evolution Algorithm (DEA).