野草算法和支持向量机的网络入侵检测

Network Intrusion Detection by Using Weed Algorithm and Support Vector Machine

为了获得更加理想的网络入侵检测结果,针对网络入侵特征选取和参数选择问题,提出一种野草算法和支持向量机的入侵检测模型。首先提取网络入侵特征,采用野草算法选择比较重要特征,然后采用最优特征训练支持向量机建立网络入侵行为识别器,并采用野草算法选择最优参数,最后采用KDD Cup99数据集进行性能测试。结果表明,本文模型得到了理想的网络入侵检测结果,检测率超过90%,入侵检测效率可以满足网络安全实际应用要求。

In order to obtain more ideal network intrusion detection results,aiming at solving the problem of selecting network intrusion features and parameters,this paper presents an intrusion detection model by using weed algorithm and support vector machine. Firstly,the features of network intrusion are extracted a selection of the important features by using the weeds algorithm,and then the optimal features are used to train support vector machine to establish network intrusion recognition,and adopts the weeds algorithm to select the optimal parameters,finally,simulation experiments is testing the performance by KDD Cup99 data set. The simulation results show that the presented model has been better network intrusion detection results,the detection rate is above 90%,the efficiency of intrusion detection can meet the requirements of network security protection.