粒子群算法和K近邻相融合的网络入侵检测

摘要

为了提高网络入侵检测效果，提出一种粒子群优化算法（PSO）和K最近邻相融（KNN）的网络入侵检测模型（PSO-KNN）。首先特征子集和KNN参数作为一个粒子，然后通过粒子之间的信息交流和相互协作，找到最优特征子集和KNN参数，从而建立最优网络入侵检测模型，最后利用KDD 1999数据集对模型性能进行测试。结果表明，相对于其他入侵检测算法，PSO-KNN更有效地精简网络数据特征，提高分类算法的网络入侵检测速度及检测率。%In order to improve network intrusion detection performance, this paper proposes a network intrusion detection model based on particle swarm optimization and k-nearest neighbor. Firstly, the features and KNN’s parameters are taken as a particle, and then the optimal features and KNN’s parameters are got by particle swarm to build the optimal network intrusion detection model, finally, the performance of the built model is tested by KDD 1999 data. Experimental results show that the proposed algorithm is more effective for feature selection of network data and improvement of network intrusion detection speed and detection rate of classification algorithms.