基于特征选择和支持向量机的入侵检测方法

摘要

In order to remove the redundancy and noise characteristics in network intrusion detection data set, reduce the data processing difficulty and improve the detection performance, a intrusion detection method based on feature selection and support vector machine is proposed. The method using the proposed feature selection algorithm base on the model of support vector machine classifier to select the best combination of features is applied to intrusion detection systems. Simulation results show that this method can not only reduce the number of features, thereby decreasing the training and testing time, but also improve the classification accuracy of intrusion detection.%为去除网络入侵数据集中的冗余和噪声特征，降低数据处理难度和提高检测性能，提出一种基于特征选择和支持向量机的入侵检测方法。该方法采用提出的特征选择算法选取最优特征组合，并以支持向量机为分类器建立模型，应用于入侵检测系统。仿真结果表明，本文方法不仅可以减少特征维数，降低训练和测试时间，还能提高入侵检测的分类准确率。