Intrusion Detection System Based on Machine Learning Algorithm to Detect the DDoS Attack in the IoT Network

Combating Distributed Denial of Service (DDoS) attacks in Internet of Things (IoT) networks is essential for enhancing network security. This paper introduces an innovative predictive model engineered to anticipate and mitigate DDoS attacks in IoT environments. By leveraging a curated subset of malicious scenarios, we utilize descriptive analytics alongside various Decision Tree (DT) algorithm models to achieve high performance with minimal complexity. Our approach involves comprehensive data collection, encompassing IoT device parameters, network traffic features, and labeled data indicating either DDoS attacks or normal behavior. Through meticulous data preprocessing, model training, and evaluation, we have assessed the efficacy of DT algorithms. Employing a cross-validation procedure, we validated the models' performance using standard confusion metrics such as accuracy, precision, recall, and F1-score. Our proposed methodology demonstrates promising results in real-time DDoS attack detection, underscoring its potential to enhance IoT network security through proactive threat mitigation. The model, developed using the IoT-Malware-Capture-60-1 dataset, achieves the highest accuracy in detecting untrained outsourced data with minimal complexity.