DDoS Attacks Detection Using Machine Learning Algorithms

Abstract

A distributed denial-of-service (DDoS) attack is a malicious attempt to disrupt normal traffic of a targeted server, service or network by overwhelming the target or its surrounding infrastructure with a flood of Internet traffic. It has caused great harm to the security of the network environment. This paper develops a novel framework called PCA-RNN (Principal Component Analysis-Recurrent Neural Network) to identify DDoS attacks. In order to comprehensively understand the network traffic, we select most network characteristics to describe the traffic. We further use the PCA algorithm to reduce the dimensions of the features in order to reduce the time complexity of detection. By applying PCA, the prediction time can be significantly reduced while most of the original information can still be contained. Data after dimensions reduction is fed into RNN to train and get detection model. Evaluation result shows that for the real dataset, PCA-RNN can achieve significant performance improvement in terms of accuracy, sensitivity, precision, and F-score compared to the several existing DDoS attacks detection methods.