An Enhanced Machine-Learning Model For Network Intrusion Detection System

Abstract

The internet and technological advancements have facilitated faster communication and information sharing. However, cybercrime, including malware, phishing, and ransomware, remains a severe problem despite technical progress. A significant challenge that has emerged with the quickening pace of technological advancement is detecting the intrusion through Intrusion Detection System (IDS) in wireless networks (WSN) and network communication. To address these challenges, this paper proposes two accurate approaches for intrusion detection in the network and WSN using machine-learning methods include Chaotic Maps (Circle and Logistic), Cauchy Mutation, Support Vector Machine (SVM), Pearson Correlation Coefficient Analysis (PCCA), and Nomadic People Optimizer (NPO). The proposed approaches have five main stages, which are data collection, pre-processing, feature selection, classification, and evaluation. Two datasets help in evaluating the proposed methods, and for WSN-DS and NSL-KDD attain accuracy 99.98 and 99.96% correspondingly.