Deep Learning Based Approach for Secure Web of Things (WoT)

Abstract

Internet of Things (IoT) includes smart devices that are connected through a common network, in order to increase the potential of these smart devices, the concept of the Web of things (WoT) has been introduced. The main aim of WoT is to connect all the smart devices through the internet so that they can share the services and resources globally. But this increase in connectivity makes the devices vulnerable to different types of cyber-attacks. Different types of cyber-attacks like DDoS attacks, DoS attacks, etc., affect the normal operation of smart devices and leak private information, so detection and prevention of cyber-attacks in the WoT is an important research issue. In this paper, we proposed a Deep-learning-based approach for the detection of different cyber attacks like DoS, U2R, R2L in the WoTs. We used the KDDCUP99 dataset for training and testing purposes and achieved an accuracy of 99.73%. We also compared our proposed approach with other machine learning approaches and check its effectiveness.