An intelligent anomaly analysis for intrusion detection based on SVM
Yong Cheng Xie, Yilai Zhang · 2012
The application of support vector machine(SVM) for network intrusion detection was researched, Although SVM was an effective abnormal analysis for intrusion detection with a small sample, there were two deficiencies in traditional SVM: slow in training, low detection rate. An intelligent anomaly analysis algorithm for intrusion detection based on SVM is presented. This algorithm can intelligently select learning vector samples during the training state, and effectively reduce the number of training samples and training time, and also can obtain a higher detection rate classifier in the case of small samples.