基于特征选择和SVM参数同步优化的网络入侵检测

为了提高网络入侵检测正确率,利用特征选择和支持向量机（SVM）参数间的相互联系,提出一种特征选择和SVM参数联同步优化的网络入侵检测算法.该算法首先将网络入侵检测正确率作为问题优化的目标函数,网络特征和SVM参数作为约束条件建立数学模型,然后通过遗传算法对数学模型进行求解,找到最优特征子集和SVM参数,最后利用KDD 1999数据集对算法性能进行测试.结果表明,相对于其他入侵检测算法,同步优化算法能够较快选择最优特征与SVM参数,有效提高了网络入侵检测正确率,加快了网络入侵检测速度.

Network intrusion detection based on simultaneous optimization of features selection and parameters of support vector machine

In order to improve network intrusion detection rate, this paper proposed a network in- trusion detection algorithm based on simultaneous optimization of feature selection and SVM pa- rameters which used the relationship between the feature selection and SVM parameters. Firstly, the network intrusion detection rate as the objection function to built mathematical model which the constraint conditions were the feature and SVM parameters. Secondly, the genetic algorithm was used to get the optimal features and SVM parameters. Lastly, the performance of the pro- posed algorithm was tested by KDD 1999 data. The results showed that the proposed algorithm could select the optimal features and SVM parameters to improve the network intrusion detection rate and detection speed compared with other network intrusion detection algorithms.