Survey Paper of Distributed Denial of Service Attacks in Cybersecurity based on Machine Learning

Abstract

Distributed Denial of Service (DDoS) attacks continue to pose a significant threat to modern cybersecurity infrastructures by disrupting network services and degrading system availability. With the rapid growth of cloud computing, IoT devices, and high-speed networks, traditional rule-based and signature-based detection mechanisms have become insufficient to handle the scale and complexity of evolving DDoS attack patterns. This survey paper provides a comprehensive review of machine learning (ML)-based approaches for detecting and mitigating DDoS attacks. It systematically analyzes various supervised learning techniques, including Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), along with their effectiveness in classifying network traffic as normal or malicious. The study examines different publicly available benchmark datasets, feature extraction methods, and preprocessing strategies used in recent research. Furthermore, it compares the performance of these models based on key evaluation metrics such as accuracy, precision, recall, and F1-score. The survey highlights that ensemble-based models, particularly Random Forest and XGBoost, consistently achieve higher detection accuracy and better generalization compared to traditional methods. In addition, the paper discusses current challenges such as data imbalance, real-time detection requirements, high computational cost, and adaptability to emerging attack vectors. Finally, it outlines future research directions, including the integration of deep learning techniques, hybrid models, and real-time deployment frameworks for enhanced DDoS detection. This survey contributes to a deeper understanding of ML-based cybersecurity solutions and provides insights for developing more robust, scalable, and efficient intrusion detection systems.