Network intrusion detection by artificial bee colony algorithm-based SVM

摘要

Network intrusion detection by artificial bee colony algorithm-based SVM is proposed in this paper, where artificial bee colony algorithm can be used to optimize the parameters of support vector machine (SVM). Artificial bee colony algorithm is a swarm-based optimization algorithm, which can be inspired by honeybee foraging behavior. There are three types of honeybees: employed bees, onlookers, and scouts in artificial bee colony algorithm. The testing results show that the network intrusion detection accuracy of artificial bee colony algorithm-based SVM is obtained as "94.8%," the network intrusion detection accuracy of SVM is obtained as "89.7%," and the network intrusion detection accuracy of artificial neural network is obtained as "81.0%." The comparison of network intrusion detection accuracy among artificial bee colony algorithm-based SVM and artificial neural network shows that the network intrusion detection accuracy of artificial bee colony algorithm-based SVM is higher than SVM or artificial neural network.