In Cloud Computing Detection of DDoS Attack Using AI-Based Ensembled Techniques

Abstract

The growing interest in cloud computing has resulted in an increase in the number of cyber-attacks counter to it. One such attack is a Distributed Denial of Service (DDoS) attack, which targets the cloud’s B/W, resources, and services in order to render them inconvenient to both their customers and cloud supplier. Because of the large volume of traffic that must be processed, machine learning classification algorithms, and data mining have been suggested to distinguish common packets from anomalous packets in order to improve efficiency. When it comes to cloud DDoS attack defense, feature selection has also been analyzed as an initiation phase that has the potential to increase classification accuracy, while simultaneously decreasing computational complexity by analyzed most significant features from the actual dataset, which is done in the time of supervised learning. In this paper, we supposed an ensemble-based multi-filter feature selection techniques with together the o/p of four different filter techniques in order to execute the best possible selection. An extensive experimental evaluation of our suggested technique was carried out used to the intrusion detection benchmark dataset, the NSL-KDD, and a decision tree classifier, among other tools. If we compare the results obtained with those obtained using other classification techniques, we can see that our suggested method successfully decreasing the No. of features from 41 to 13, and it has classification accuracy with high detection rate.