Real-Time DDoS Attack Detection System using Machine Learning and Network Traffic Monitoring

Distributed Denial of Service attacks (DDoS) are very serious threats in the field of cybersecurity and can overwhelm the systems and services reliant on these. It brings a real-time DDoS detection system together with live traffic monitoring and techniques of machine learning for effective identification and mitigation of such attacks. The system captures and analyzes network traffic using Scapy and extracts the significant flow-based features that include packet rates, byte rates, protocol distributions, and flow duration. This is scaled and fed into a pre-trained Logistic Regression model to classify the network activity with high accuracy as either normal or malicious. It is designed with adaptability in that it monitors the traffic patterns over configurable time windows to dynamically respond to changes in conditions. This would result in reporting in case of detection, where information such as source & destination internet protocol addresses, traffic rates, and protocol usage is provided, thereby enabling rapid response and effective mitigation strategies. The system automatically detects DDoS and thus reduces the involvement of human intervention and increases response time while strengthening overall network defenses.