Paper Contents
Abstract
As the Internet of Things (IoT) continues to proliferate, ensuring the security of connected devices becomes critical due to their heterogeneity, limited resources, and often lax configurations. This paper presents a machine learning-based framework to evaluate and assign a dynamic cybersecurity risk score to IoT devices. The proposed system considers device behavior (traffic patterns, access anomalies) and configuration parameters (default credentials, open ports, outdated firmware) to classify threat levels in real time. By employing supervised and unsupervised learning models, we demonstrate the efficacy of risk scoring in prioritizing response efforts and optimizing resource allocation. Experimental validation using benchmark datasets and simulated IoT environments shows a significant improvement in early threat detection and response agility
Copyright
Copyright © 2025 Meenuga Pranaya Praharshitha. This is an open access article distributed under the Creative Commons Attribution License.
