Experimenting Ensemble Machine Learning for DDoS Classification: Timely Detection of DDoS Using Large Scale Dataset

Abstract

The rapid expansion of the internet has connected the world with a single network. Every network is the victim of a hacker and can be attacked by finding its vulnerabilities. Distributed Daniel of Service (DDoS) attack overwhelms a network and restricts its user from accessing reachable resources. In this study, we aim to employ ensemble ML techniques, such as random forest, histogram-based gradient boosting, and adaptive boosting classifiers, to detect DDoS attacks using the CIC-DDoS2019 dataset. The comparative evaluation results of this study reveal that it provides a higher detection accuracy score (99.9887%) compared to the previous studies.