Detection of HTTP Attacks using Machine Learning

Distributed denial of service (DDoS) attacks are considered to be one of the most occurred attacks recently. Internet service and cloud providers suffer from various types of DDoS attacks. Such an attack causes the service unreachable or unresponsive to its legitimate clients. HTTP is a type of volumetric DDoS attack in which an adversary sends a massive number of HTTP requests to the victim's webserver, consuming all available resources and making the webserver unavailable to its intended clients. This article proposed a detection model to successfully predict HTTP DDoS attacks. Our proposed model employed a set of machine learning approaches for DDoS attack detection and classification. The CIDDS-001 dataset is used in our experimental results. The evaluation metrics proved that the KNN and random forest algorithms obtained higher degree of accuracy result compared with the other machine learning methods.