An Al-Driven System For DDoS Detection And Forecasting
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Distributed denial of service (DDoS) attacks, which can overload servers and networks and prevent authorized users from accessing essential services, are one of the most dangerous threats to modern digital infrastructures. These attacks cost companies money and damage their reputation in addition to causing downtime. To address this growing problem, the proposed study offers an artificial intelligence (AI)-powered framework for detecting and forecasting DDoS attacks using state-of-the-art machine learning techniques. The UNSW-NB15 dataset, which represents contemporary attack behaviors and provides a more accurate benchmark than earlier datasets, is used in the study. Two ensemble-based algorithms, Random Forest and XGBoost, were employed to classify and predict DDoS traffic. Random Forest’s accuracy of 89% was marginally lower than XGBoost’s 90%. The experimental results demonstrate that ensemble learning techniques are effective in distinguishing malicious DDoS traffic from legitimate network activity, achieving high accuracy and robust generalization across attack patterns. Crucially, the framework incorporates predictive analytics to enable the forecasting of attack likelihood, going beyond detection. Organizations are able to foresee threats and fortify defenses prior to disruptions thanks to this proactive approach. The results demonstrate how AI-driven models can improve cybersecurity resilience and provide a promising path for further study in predictive defense tactics.
Fields of Study & Subjects
Network Security and Intrusion Detection
Internet Traffic Analysis and Secure E-voting
Network Packet Processing and Optimization
Publication details
- Journal / source
- International Conference on Computing, Communication, Control and Cyber-Physical Systems (I5CPS)
- Pages
- 1–5
- Publication date
- 26 February 2026
- Language
- EN