A novel approach to intrusion detection based on SVD and SVM

This paper describes a new intrusion detection methods based on singular value decomposition and support vector machine. The proposed method utilizes a new feature based on orthogonal projection coefficients obtained by singular value decomposition. The support vector machine classifier is performed on the new extracted feature vector sets. The RBF kernel parameters are optimized by the grid-search using cross-validation in this paper. Finally experiment results show that the novel intrusion detection method is effective and possesses several desirable properties when it compared with many existing methods.