Role of Neural Network, Fuzzy, and IoT in Integrating Artificial Intelligence as a Cyber Security System

Abstract

The "Internet of Things" has a vast number interconnected devices. These interconnected devices collect vital data that may have a significant effect on the company, society, and the environment as a whole. IoT application has grown significantly in recent times, and with it, so do worries about cybersecurity. Artificial intelligence (AI) is at the forefront of the technology of cybersecurity and is employed to create intricate algorithms to safeguard systems and networks like IoT devices. But cybercriminals have learned how to take advantage of AI, and they have even started to deploy AI in analyzing cyberattacks. Due to the limited computing power and memory capacities of IoT systems, conventional high-end cybersecurity measures are inadequate to protect an IoT system. The need for accessible, distributed, and robust smart security systems is highlighted by this. Large- and small-scale heterogeneous datasets are no match for DL. In this research, a multilayer cybersecurity strategy based on DL is used to safeguard the TL of IoT systems. The developed framework tests the proposed multi-layer strategy using the intrusion identification statistics obtained from CIC-IDS (2018), ToN, and BoT-IoT. Consequently, depending on the analyzed parameters, the proposed model has outperformed the other approaches and achieved 98% accuracy.