论文部分内容阅读
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 the selection of network intrusion features and parameter selection, a weed detection algorithm based on Weeds algorithm and support vector machine is proposed. Firstly, the characteristics of network intrusion are extracted, the weeds algorithm is used to select the most important features, and then the optimal feature training SVM is used to establish the network intrusion behavior identifier. The optimal parameters are selected by the weed algorithm. Finally, the performance is tested by KDD Cup99 dataset. The results show that the model has got the ideal network intrusion detection results, the detection rate is over 90%, intrusion detection efficiency can meet the practical application of network security requirements.