Cloud-Based Intrusion Detection System Using a Deep Neural Network and Human-in-the-Loop Decision Making
Hootan Alavizadeh, Hooman Alavizadeh · 2023
In this chapter, we propose a cloud-based intrusion detection system (IDS) framework that utilizes a convolutional neural network (CNN) to detect various types of attacks. The proposed IDS framework is equipped with the human-in-the-loop (HITL) module to enhance the capabilities of our proposed cloud-based IDS based on the expert&s;s intervention and feedback. Moreover, we propose a non-zero-sum game model for the HITL module to evaluate the proposed model based on different attack scenarios that may be able to deceive the IDS system. We show that our proposed CNN model is able to detect five different attack classes with high accuracy. Moreover, our CNN model outperforms other related approaches in the literature.