Application of Support Vector Machine and Genetic Algorithm to Network Intrusion Detection
Hua Zhou, Xiangru Meng, Li Zhang · 2007
Intrusion detection is actually a classification problem. It is very important to increase the classification accuracy. Support Vector Machine (SVM) is a powerful tool to solve classification problems. Many works have been done in intrusion detection based on SVM, and the detection accuracy is relatively high. But how to get a higher accuracy is a new question. In this paper, we apply SVM and Genetic Algorithm (GA) to intrusion detection to solve this problem. We first use GA for feature selection and optimization, and then use SVM model to detect intrusions. In order to verify our approach, we tested our proposal with KDD Cup99 dataset, and analyzed its performance. The experimental results show that the proposed approach is an efficient way in network intrusion detection.