Development of Hybrid Deep Network-based Decision Making Process for Cognitive Cyber Security by Cognitive Science

The development of present technology such as social networks, mobile computing, cloud computing, and Internet of Things (IoT) faces security attacks and threats. This tends to generate a new cybersecurity scenario by evaluating the huge volume of data along with the multiple attack vectors. However, previous research work in cybersecurity faces various complex issues in securing interconnected and dynamic Cyber-Physical Systems (CPSs). Further, the existing approaches employ the human-like cognitive mechanism, which faces difficulties in performing various tasks like decision-making, learning, and control functions. Hence, a new cognitive cyber security model is implemented in this work by leveraging cognitive science. At first, the required data is gathered from the various datasets and then it is subjected to the Hybrid Network (HyNet) for decision-making. Here, techniques such as Efficient Capsule Network (ECNet) and Bayesian Learning (BL) are serially cascaded to provide accurate decisions in cyber security. Finally, the developed HyNet offers a decision making outcome. The implemented model is compared with the traditional models to analyze the effectiveness of the model.