Research on intrusion detection based on improved combination of K-means and multi-level SVM

Aiming at the problem that the traditional network intrusion detection algorithm has the advantages of low detection efficiency and high false alarm rate, a network intrusion detection algorithm based on improved K-means and multi-level SVM is proposed. The algorithm first divides the data to be detected into different clusters with the improved K-means, and marked as normal or abnormal; and then use the multi-level SVM to mark the abnormal cluster for detailed classification, the final realization of the detection of network attacks. The proposed intrusion detection algorithm uses the NSL-KDD data set to simulate the experiment. The results show that the proposed algorithm can improve the network intrusion detection rate and reduce the false alarm rate. It is an effective way of network security protection.