Design of Threat Detection Model for IoT-Enabled Smart Environment
Anchal Dahiya, Pooja Mittal · 2023
In an era marked by the pervasive integration of Internet of Things (IoT) technology into diverse sectors, safeguarding the security of IoT-enabled smart environments has emerged as a paramount concern. This paper presents a threat detection model specifically designed for IoT-enabled smart environments. A “DS2OS” dataset is utilized for evaluation, and machine learning techniques, including SVM, DT, RFT, and XGBT, are employed. The results show high accuracy, precision, and recall for each technique. Furthermore, a proposed voting classifier achieves exceptional performance with a 99% accuracy rate, contributing to improved security and safety in IoT-enabled environments. By providing an advanced threat detection mechanism, this model contributes to enhancing the security and reliability, ultimately promoting the widespread adoption and utilization of IoT technology for societal benefit.