Artificial Intelligence-Based Anomalies Detection Scheme for Identifying Cyber Threat on IoT-Based Transport Network

Increasing use of portable wireless devices in the Internet of Things (IoT) network has made it more dynamic, flexible, and vulnerable to cyber-attacks due to shared communication links, and it is critical to identify and mitigate potential security risks. Thus, this leads to the crucial need for an intrusion detection system that can uncover malicious attacks on the IoT network. In order to identify malicious sessions in IoT networks, the author proposes an artificial intelligence-based IDS model employing a feature selection technique based on fuzzy and genetic algorithms (GA). We use the bio-inspired genetic algorithms Intelligent Water Drop (IWD) and Biogeography-based Optimization (BBO) for feature selection. We provide an effective feature extractor that employs intelligent water drop (IWD) algorithms and a feed-forward network called the fuzzy water drop intrusion detection model (FWDNN) for assault categorization. In this paper, we propose an artificial intelligence-based IDS model using feature selection method based on fuzzy and genetic algorithms (GA) with the goal of detecting malicious sessions in IoT networks. Evaluation is done on real IoT datasets and CICIDS-2017, and the results show that the proposed BBOKNN model outperforms existing models in terms of evaluation parameters.