当前位置: X-MOL 学术Neural Comput. & Applic. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A stacked ensemble learning model for intrusion detection in wireless network
Neural Computing and Applications ( IF 6 ) Pub Date : 2020-05-21 , DOI: 10.1007/s00521-020-04986-5
Hariharan Rajadurai , Usha Devi Gandhi

Intrusion detection pretended to be a major technique for revealing the attacks and guarantee the security on the network. As the data increases tremendously every year on the Internet, a single algorithm is not sufficient for the network security. Because, deploying a single learning approach may suffer from statistical, computational and representational issues. To eliminate these issues, this paper combines multiple machine learning algorithms called stacked ensemble learning, to detect the attacks in a better manner than conventional learning, where a single algorithm is used to identify the attacks. The stacked ensemble system has been taken the benchmark data set, NSL-KDD, to compare its performance with other popular machine learning algorithms such as ANN, CART, random forest, SVM and other machine learning methods proposed by researchers. The experimental results show that stacked ensemble learning is a proper technique for classifying attacks than other existing methods. And also, the proposed system shows better accuracy compare to other intrusion detection models.



中文翻译:

无线网络中入侵检测的堆叠集成学习模型

入侵检测被认为是揭示攻击并保证网络安全的主要技术。由于Internet上的数据每年都在以惊人的速度增长,因此单一算法不足以保证网络安全。因为部署单一学习方法可能会遇到统计,计算和表示问题。为了消除这些问题,本文结合了称为堆叠集成学习的多种机器学习算法,以比传统学习更好的方式检测攻击,传统学习使用一种算法来识别攻击。堆叠集成系统已采用基准数据集NSL-KDD,以将其性能与其他流行的机器学习算法(如ANN,CART,随机森林,SVM和研究人员提出的其他机器学习方法)进行比较。实验结果表明,与其他现有方法相比,堆叠集成学习是一种对攻击进行分类的合适技术。而且,与其他入侵检测模型相比,该系统显示出更好的准确性。

更新日期:2020-05-21
down
wechat
bug