Performance Analysis of Classification Algorithms for Distributed Denial of Service Attacks Detection in a Distributed Network Environment

Abstract

Organization network and its infrastructures persistently face challenges of Distributed Denial of Service (DDoS) attacks [19]. Mostly the attacks are targeted at the crucial network infrastructures such as the database server, cloud computing server, web server and other computing devices. The occurrence of such attacks causes a serious negative impact to the organization and its vital infrastructures. In this paper, six well-known classification algorithms (Random Forest, Decision Stump, NNge, OneR, RART and Naïve Bayes algorithms) were applied on NSL-KDD dataset to examine the performance of individual algorithm in terms of accuracy and false detection rate. The dataset was streamlined for optimum performance of the selected algorithms. The experimental result shows that Random Forest algorithm has 98.7% Detection accuracy and false detection rate of 0.022%.