ARCHIVES

The exponential growth of internet infrastructure has intensified the threat landscape for network-based services, with Distributed Denial-of-Service (DDoS) attacks emerging as one of the most destructive cybersecurity challenges. These attacks overwhelm network resources with massive volumes of malicious traffic, rendering services unavailable to legitimate users and causing financial and reputational damage. This paper proposes an intelligent framework for detecting and mitigating DDoS attacks using machine learning techniques. The system employs the Random Forest Classifier and Random Forest Regressor algorithms trained on network traffic features including flow duration, packet size, and flow bytes per second, CPU utilization, memory utilization, and anomaly scores. The framework classifies incoming traffic as Normal Traffic or DDoS Attack in real time, while simultaneously estimating the anomaly severity index and risk level. Upon detecting malicious activity, the system automatically triggers mitigation actions such as blocking suspicious IP addresses and applying rate limiting. The web-based interface is developed using HTML, CSS, JavaScript, and Flask framework. Experimental results demonstrate high detection accuracy, low false positive rates, and robust real-time response performance, making the proposed framework a reliable solution for modern network intrusion detection and prevention.