Research of Intrusion Detection Based on Support Vector Machine

摘要

For the network data sets too large, too slow learning speed problem, in this paper, a SVM algorithm based on space block and sample density is proposed and applied into intrusion and detection. According to the local density the algorithm selects training samples and reduces the number of training sample to enhance learning speed. The algorithm can guarantee the accuracy of detection and at the same time the learning speed of it is faster than the traditional SVM intrusion detection method.