Publication Details

Abstract

This research talks about how smart gadgets that are connected to the internet make more attacks possible. It concentrates on a cybersecurity design that is based on machine learningS and is intended to safeguard Internet of Things (IoT) networks. The proposed framework employs sophisticated machine learning techniques, including anomaly detection algorithms and classification algorithms, to identify and resolve cyber hazards in real time. The system enhances the precision of threat identification and reduces the number of false positives by analysing network traffic patterns, device behaviour, and potential intrusion signatures. The framework employs adaptive learning techniques to adapt to new attack vectors and ensure that IoT settings that are constantly evolving are secure and scalable. This approach appears to be a viable substitute for conventional rule-based security systems in terms of safeguarding existing IoT infrastructures. It is more effective at detection, responds more quickly, and offers superior protection, as demonstrated by experiments.