Cyber intrusion detection by combined feature selection algorithm
Cyber intrusion detection by combined feature selection algorithm
复制标题
DOI：
10.1016/j.jisa.2018.11.007
复制
发表时间：
2019-02-01
影响因子：
5.6
通讯作者：
Karimipour, Hadis
中科院分区：
文献类型：
--
作者：
Mohammadi, Sara;Mirvaziri, Hamid;Karimipour, Hadis
由于网络连接的广泛普及，对网络安全和网络攻击防护的需求不断增加。入侵检测系统在当今的网络安全中起着至关重要的作用。提出了一种基于特征选择和基于过滤器和包装器的聚类算法的入侵检测系统。滤波法和包装法分别称为基于线性相关系数的特征分组(FGLCC)算法和基于墨鱼的特征分组(CFA)算法。该方法采用决策树作为分类器。为了验证该方法的性能，在KDD CUP 99大数据集上进行了应用。结果表明，与现有方法相比，该方法具有较高的准确率(95.03%)和较高的检测率(95.23%)，较低的假阳性率(1.65%)。(C)2018爱思唯尔有限公司。保留所有权利。
Due to the widespread diffusion of network connectivity, the demand for network security and protection against cyber-attacks is ever increasing. Intrusion detection systems (IDS) perform an essential role in today's network security. This paper proposes an IDS based on feature selection and clustering algorithm using filter and wrapper methods. Filter and wrapper methods are named feature grouping based on linear correlation coefficient (FGLCC) algorithm and cuttlefish algorithm (CFA), respectively. Decision tree is used as the classifier in the proposed method. For performance verification, the proposed method was applied on KDD Cup 99 large data sets. The results verified a high accuracy (95.03%) and detection rate (95.23%) with a low false positive rate (1.65%) compared to the existing methods in the literature. (C) 2018 Elsevier Ltd. All rights reserved.