文章基本信息

标题：A Study of DDoS Attacks Detection Using Supervised Machine Learning and a Comparative Cross-Validation
本地全文：下载
作者：Wedad Alawad ; Mohamed Zohdy ; Debatosh Debnath 等
期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：12
页码：17659
DOI：10.15680/IJIRCCE.2017.0512068
出版社：S&S Publications
摘要：Despite the benefits that encourage organizations to move toward the cloud, security issues are strongbarriers. Distributed denial of service (DDoS) attack can target cloud computing environments and compromise theavailability of cloud-based services. Thus, offensive techniques are highly recommended to detect DDoS and decreasethe possibility of their success. One of the techniques used to detect such attacks is machine-learning. In this paper, theperformance and detection accuracies of three supervised machine learning classifiers are compared:Naive Bayes,Decision Tree, and Linear Discriminate Analysis. The impact of the training sample size onclassifier accuracy isinvestigated as well. Furthermore, a novel accuracy estimation method, F-Hold Cross-Validation, is proposed andcompared to the K-Fold Cross-Validation method to assess it. The results show that F-Hold Cross-Validation is timeefficientand its estimated values are acceptable.
关键词：Machine learning; security; cloud computing; Cross-Validation; supervised classifier