Paper Title
Machine Learning Models for the Detection of DDOS Attacks Using BOTNETS

Abstract
Botnet-based Distributed Denial of Service (DDoS) attacks pose a significant threat to network availability and cybersecurity due to their dynamic and evolving attack patterns. Traditional detection approaches are often ineffective in identifying modern large-scale DDoS attacks. In this paper, a hybrid detection framework named SVM-RNNbGA is proposed, which integrates Support Vector Machines (SVM), Recurrent Neural Networks (RNN), and Genetic Algorithms (GA) for intelligent botnet-based DDoS attack detection. The SVM component performs efficient traffic classification, while the RNN captures temporal traffic behaviour. Furthermore, the Genetic Algorithm optimizes model parameters to improve detection performance and reduce false positive rates. The proposed framework was evaluated using the CICDDoS2019 dataset. Experimental results demonstrate that the proposed SVM-RNNbGA model achieves an accuracy of 99.85%, precision of 99.20%, recall of 99.35%, and F1-score of 99.25%. The results indicate that the proposed framework provides reliable, balanced, and efficient DDoS attack detection with improved overall system performance. Keywords Machine Learning, Deep Learning, DDoS Detection, Botnet Detection, Genetic Algorithm, Cybersecurity