Robust Intrusion Detection Algorithm Based on K-Means and BP
Nowadays the traditional intrusion detection models have two disadvantages:low work efficiency and high false positive.Considering these disadvantages,this paper proposes a new intrusion detection method combining k-means clustering algorithm and BP neural network.The experimental results show that the improved intrusion detection model in this paper can save the calculation time significantly with same detection capabilities for the abnormal behavior.
Shaobo Zhong, Yimin Cheng and Xilong Qu
Y. J. Zhong and S. P. Zhang, "Robust Intrusion Detection Algorithm Based on K-Means and BP", Applied Mechanics and Materials, Vols. 50-51, pp. 634-638, 2011