A Comparative Study of Machine Learning and Deep Learning Approaches for Enhancing Intrusion Detection in IOT Environments

Abstract

With the increasing deployment of IOT and Edge Cloud environments, ensuring the security of these systems against cyber-attacks bas become a critical challenge. Traditional intrusion detection systems (IDS) often fall short in addressing the complexity and scale of attacks in such heterogeneous environments. This paper aims to explore machine learning (ML) and deep learning (DL) techniques for efficient attack detection in IOT environments, with a focus on evaluating the performance of various models. The objective was to analyses the effectiveness of model, i.e., XGBoost (XGB) in identifying attacks using the TON-IoT dataset. The findings showed that XGB achieved an accuracy of 95.04%. The paper concludes that advanced ML and DL models, especially ensemble methods, offer significant improvements in intrusion detection. The novelty of this work lies in comparing various state-of-the-art approaches on a insights for enhancing security in Edge Cloud environments.