DDoS Attack Detection

Abstract

DDoS (Distributed Denial of Service) attacks are becoming a real problem for cloud servers and networks, causing major disruptions and financial losses. As technology keeps evolving, Software Defined Networks (SDN) and Machine Learning (ML) are offering new ways to fight back. In this paper, we introduce a machine learning model that can quickly adapt to new types of DDoS attacks as they happen. The model focuses on the most important data, which makes it really good at catching attacks. Our results were pretty impressive, with the model detecting 99.2% of attacks, outperforming others on well-known datasets like UNSW_NB15 and In SDN, We also looked into how the Random Forest algorithm can detect and block DDoS attacks in real-time, especially in cloud systems, by analyzing network traffic. This research shows how effective machine learning can be in improving network security and reducing the harm caused by DDoS attacks. Ultimately, our work offers a strong solution for protecting SDN systems and ensuring cloud services remain safe from the rising threat of cyberattacks.

KEYWORDS: Distributed Denial of Service (DDoS) Detection, Cloud Computing Security, Software Defined Networking (SDN), Machine Learning Algorithms, Random Forest Classifier, Intrusion Detection System (IDS), UNSW_NB15 Dataset, Network Traffic Analysis