Detecting Distributed Denial of Service Attacks Using Data Mining Techniques

TLDR

Distributed denial‑of‑service attacks continuously challenge users and organizations, requiring security engineers to maintain service availability through updated intrusion‑detection systems that detect and classify anomalous behavior. The study aims to create a new dataset encompassing modern DDoS attacks across multiple network layers, such as SIDDoS and HTTP flood. The authors evaluate three classifiers—Multilayer Perceptron, Naïve Bayes, and Random Forest—on the newly collected dataset. Multilayer Perceptron achieved the highest accuracy of 98.63%, outperforming Naïve Bayes and Random Forest. No additional metadata provided.

Abstract

Users and organizations find it continuously challenging to deal with distributed denial of service (DDoS) attacks. . The security engineer works to keep a service available at all times by dealing with intruder attacks. The intrusion-detection system (IDS) is one of the solutions to detecting and classifying any anomalous behavior. The IDS system should always be updated with the latest intruder attack deterrents to preserve the confidentiality, integrity and availability of the service. In this paper, a new dataset is collected because there were no common data sets that contain modern DDoS attacks in different network layers, such as (SIDDoS, HTTP Flood). This work incorporates three well-known classification techniques: Multilayer Perceptron (MLP), Naïve Bayes and Random Forest. The experimental results show that MLP achieved the highest accuracy rate (98.63%).