Network intrusion detection system based on Direct LDA

Within the last years, lot of cyber attacks target sensitive governmental and bank data, therefore IT security has become a major economic issue. Intrusion detection systems are an effective way to detect malicious connections in a computer network. To get there, they classifies the connections as normal or abnormal, using different algorithms. In this paper three methods are used, Principal Component Analysis (PCA) and the two-stage algorithm PCA+LDA as related works, and the Direct Linear Discriminant analysis (D-LDA) algorithm as a proposed method. The PCA step in the PCA+LDA algorithm eliminates the problem of singularity but the PCA criterion is not compatible with the LDA criterion and thus may discard the important discriminative information, unlike the D-LDA algorithm which keeps them and gets rid of the singularity problem. In this work, we propose its implementation for intrusion detection system. Experimental results show that the proposed method is promising in terms of detection accuracy rate which is 97.75%.