A Critical Review on Learning Behavior for Protection of User's Privacy using IOT

Abstract

Changes and improvements to human existence have been made possible by recent advancements in communication and information technology, notably the internet of things (IoT). The IoT system is vulnerable to cyber-physical security and privacy assaults such denial of service, spoofing, phishing, obfuscations, and jamming because of the widespread availability and rising demand for smart devices. Cyber dangers to IoT systems, such as eavesdropping, attacks, and more. The new threats to cyber-physical security cannot be effectively avoided or mitigated using the same old methods. Keeping IoT systems safe calls on security measures that are not only effective, but also flexible and up-to-date. Among the various approaches to cyber-physical system security, machine learning (ML) is widely regarded as the most cutting-edge and promising since it has spawned several new lines of inquiry into the problem (CPS). This literature study provides an overview of the structure of Internet of Things (IoT) systems, explores the many attacks that may be launched against them, and discusses the current thinking on how to best use machine learning to ensure the security and safety of IoT infrastructure. It also covers the probable future research obstacles that may arise while implementing security measures in IoT systems.