An intrusion detection method based on KICA and SVM

Recently, support vector machine (SVM) has become a popular tool in classification, feature extraction is an important step in developing a successful classifier. In this paper, a novel intrusion detection method based on KICA and SVM is proposed. In the proposed method, KICA is applied to extraction features from the raw data set captured from the network, and these features extracted by KICA is used as input data of SVM, which can learn from the input data. Based on the good performance of SVM in generalization, experimental results show that this model can not only detect existed attacks but also new attacks, even the accuracy is improved remarkably.

[1]  Francis Eng Hock Tay,et al.  A comparative study of saliency analysis and genetic algorithm for feature selection in support vector machines , 2001, Intell. Data Anal..

[2]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[3]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[4]  D.S. Bauer,et al.  NIDX-an expert system for real-time network intrusion detection , 1988, [1988] Proceedings. Computer Networking Symposium.

[5]  Alexander J. Smola,et al.  Learning with kernels , 1998 .

[6]  Shigeo Abe,et al.  Fuzzy least squares support vector machines for multiclass problems , 2003, Neural Networks.

[7]  Michael I. Jordan,et al.  Kernel independent component analysis , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[8]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[9]  Michael I. Jordan,et al.  Kernel independent component analysis , 2003 .

[10]  Marc Dacier,et al.  Towards a taxonomy of intrusion-detection systems , 1999, Comput. Networks.