DDoS Attacks Detection in IoT Networks using Naive Bayes and Random Forest

The proliferation of Internet of Things (IoT) devices has resulted in numerous benefits, including streamlined tasks and improved connectivity but has also raised concerns regarding security vulnerabilities. Distributed Denial of Service (DDoS) attacks targeting IoT systems pose a significant threat, potentially disrupting operational activities. This study conducts a comparative analysis of existing methodologies to detect DDoS attacks in IoT environments, focusing on the efficacy of machine learning algorithms. Specifically, this study evaluates the performance of Random Forest and Naive Bayes classifiers in detecting DDoS attacks using an IoT dataset. The evaluation involves training and testing these classifiers and assessing their accuracy as a key performance metric. Empirical findings demonstrate enhanced precision in identifying DDoS attacks in IoT environment, indicating the effectiveness of the proposed approach. By leveraging machine learning algorithms, particularly Random Forest and Naive Bayes, traditional techniques for detecting and mitigating DDoS attacks in IoT environments are accelerated, thus enhancing the security of IoT systems.