Detecting DDoS Attacks on Network Traffic Using a Hybrid Long Short-Term Memory (LSTM) and Support Vector Machine (SVM) Algorithm

Abstract

Distributed Denial of Service (DDoS) the attack had a significant impact risk for network security by flooding systems with excessive traffic, disrupting services, and causing potential financial harm [1]. As these attacks grow more frequent and sophisticated, effective detection methods are essential [2]. Machine learning techniques offer a powerful solution by identifying abnormal traffic patterns associated with DDoS attacks. This study focuses on developing a detection model that combines LSTM and SVM algorithms [3]. LSTM component analyzes time-based traffic trends, while the SVM distinguishes between normal and malicious activity [4]. Performance is assessed using metrics accuracy, precision, recall, and F1-score. This study shows that the hybrid LSTM-SVM model performs very well, achieving 95% accuracy, 91% precision, 96% recall, and 93% F1-score. These results highlight the model's potential as a powerful tool for improving DDoS attack detection and strengthening network security defenses.