A Review on Optimization and Feature Selection Techniques for Data Security in IoT

Since IoT devices are becoming increasingly widespread, the security of IoT-dependent networks gets highly significant. Because of the Internet of Things' fast global proliferation, there has been a massive surge in cyber-attacks on it. Such products are connected to end-user over a public platform for flexibility of use, thereby increasing the scope of threat and endangering the confidentiality of users' data. Anyone may utilize the flaw in IoT nodes virtually anywhere that emphasizes processing resources confidentiality and safety. Optimization algorithms with the strongest feature selection approach were proposed as a potential option for minimizing different threats and to ensure data security. This paper examines and reviews Machine Learning and Evolutionary Swarm Intelligence based Optimization Algorithms with existing feature selection techniques. Further it discusses the significance of these techniques in IoT security and gives a thorough comparison of several optimization models along with the feature selection strategies used to build the models.