Comparative Analysis to Identify the Effective Machine Learning Method for Prediction of DDOS Attack

DDOS is a critical and common attack that affects the performance and reliability of the network. In this attack, the network traffic is increased anonymously on a particular node or controller. These heavy communications occupies the resources and slow down the network performance. This attack is more challenging for Cloud and IoT based networks. The lightweight components increase the network and communication criticality over the network. Various machine learning methods and models are available to predict DDOS attack. In this paper, a comparative analysis is provided on five machine learning algorithms called decision tree, decision table, random forest, SVM and Bayesian network algorithms. The analysis of these algorithms is provided using accuracy, True positive, true negative, precision and recall parameters. The analysis results show that the decision tree achieved the better performance over other algorithms. The results also show that the Bayesian network is the poor performer as compare to others.