【摘要】 为了提高网络入侵检测效果以加强网络安全性,提出一种网络状态特征和支持向量机(SVM)参数联合选择的网络入侵检测模型(PSO-SVM)。以网络入侵检测正确率作为目标,特征子集和SVM参数作为约束条件建立数学模型,通过粒子群优化算法对模型进行求解,找到最优特征子集和SVM参数,利用KDD Cup 99数据集对算法性能进行测试。测试结果表明,相对于其它入侵检测算法,PSO-SVM可以找到更优特征子集和SVM参数,加快了检测速度,有效地提高了网络入侵检测正确率,为网络入侵检测提供了一种新的研究思路。更多还原

【Abstract】 In order to improve network intrusion detection rate,a network intrusion detection model(PSO-SVM) based on jointly selection of support vector machine(SVM) parameters and features selection is proposed.Firstly,the network intrusion detection rate is taken as the objection function to built mathematical model which the constraint condition is the optimal features and SVM parameters,secondly,the particle swarm optimization algorithm is used to solve the mathematical model to get the optimal features and SVM parameters,lastly,the performance of the jointly optimization algorithm is tested by KDD Cup 99 data.The results show that the proposed algorithm can quickly select the optimal features and SVM parameters to improve the network intrusion detection speed and detection rate compared with other algorithms,and a new research way for network intrusion detection is provided.更多还原