An Algorithm for K-LVQ Abnormal Traffic Classification Based on Information Entropy

Previous researches on abnormal traffic detection mainly focus on how to detect outliers. For some detection methods based on classification, most of the anomaly classification is simultaneous detection and detection, and most of the samples are normal traffic. On the one hand, the classification accuracy is not high, on the other hand, it is not conducive to identify the specific types of abnormal traffic. For this reason, a method of classifying and identifying abnormal traffic is proposed: K-LVQ classification algorithm based on information entropy. On the basis of quantifying the entropy of each attribute of abnormal flow, firstly clustering and locating the abnormal sample points, and then learning the vector quantization neural network (LVQ) method to determine the recognition. Make full use of K-Means efficient and LVQ accurate features. The simulation results and measured data show that the average accuracy of classification and recognition can reach more than 89%.