Intrusion Detection System by Integrating PCNN and Online Robust SVM
Hengjie Li, Jiankun Wang · 2007 IFIP International Conference on Network and Parallel Computing Workshops (NPC 2007) · 2007
This paper proposes the application of principal component neural networks for intrusion feature extractions, the extracted features are employed by online robust SVM for classification. The MIT's KDD Cup 99 dataset is used to evaluate the proposed method compared to conventional SVMs, ANN and KNN in separating normal usage profiles from intrusive profiles of computer programs, which clearly demonstrates that PCNN-based feature extraction method can greatly reduce the dimension of input space without degrading or even boosting the classification performance, and indicates the superiority of online Robust SVM not only can achieve high intrusion detection accuracy and low false positives but also can be trained online and the results outperform the original ones with fewer support vectors and less training time without decreasing detection accuracy. Both of these achievements could significantly benefit an effective online intrusion detection system.