DDOS DETECTION AND MITIGATION SYSTEMS USING MACHINE LEARNING

ABSTRACT This research presents the development of a DDoS Detection and Mitigation Model using Machine Learning, aimed at enhancing cybersecurity defenses against evolving Distributed Denial-of-Service (DDoS) attacks. Traditional rule-based detection methods often struggle with adaptability and accuracy, necessitating the integration of machine learning techniques to improve real-time attack identification. The study involves data collection from Maggie, feature selection, model training and performance evaluation using machine learning algorithms such as Random Forest, Logistic Regression and Decision Tree models. The proposed model effectively analyses network traffic patterns, distinguishing between legitimate and malicious requests while reducing false positives. Evaluation metrics including accuracy, precision, recall, F1-score, and ROC-AUC curve confirm the model’s efficiency in detecting DDoS threats with high reliability. Security measures such as data encryption, anomaly detection logs and continuous retraining with new attack data further enhance its robustness. The findings highlight the model’s scalability, adaptability and potential integration into Intrusion Detection Systems (IDS) and cloud-based security frameworks. Future recommendations include incorporating deep learning techniques, AI-driven automated mitigation strategies, and blockchain-based authentication to further enhance network resilience against sophisticated cyber threats