Application of Machine Learning in Malicious IoT Classification and Detection on Fog-IoT Architecture

Due to the limitations in self-protection and information processing capabilities at IoT (Internet of Things) nodes, these nodes are susceptible to attacks, turning them into malicious nodes that cause damage or danger to the system. Early detection of these threats is essential to make timely recommendations and limit severe consequences for individuals and organizations. The study proposes applying a machine learning model to detect malicious traffic and IoT devices, which can be deployed and applied on the Fog IoT platform. This solution helps detect and early warn threats from IoT data before they are sent to the cloud. The model is evaluated on the IoT-23 dataset and gives good results.

---

Opracowanie rekordu ze środków MNiSW, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).