ANOMALY DETECTION IN IOT NETWORK USING DEEP NEURAL NETWORKS IT-801

Abstract

Internet of things (IoT) is a continuous flow of information among many small power embedded machines that work based on the Internet to link with one another. It is anticipated that the IoT will be extensively installed and will find usage in numerous areas of life. The need for IoT has recently engrossed vast attention, and administrations are keen about the data generated from the devices by arraying such systems. On the conflicting side, IoT has several safety and confidentiality worries for the users that put a break on its rise. Outbreak and irregularity uncovering in the Internet of Things (IoT) setup are some of the mounting concerns in the area of IoT. With bigger usage of the Internet of things(IoT) everywhere, threats and outbreaks are also increasing. In our paper, we have used machine learning and deep neural models to detect different kinds of outbreaks and abnormalities on IoT machines. Results were based on the estimation of efficiency given by precision, accuracy, f1 score, recall, and ROC Curve. Our model resulted in 99.43% accuracy using a deep neural network.