Machine Learning-based Intrusion Detection Systems for Securing Cloud Networks

Abstract

  Although cloud computing has become widely used, it has brought up a number of security concerns in addition to greatly improving resource economy and accessibility. The purpose of this study is to find out how well intrusion detection systems (IDS) based on machine learning can improve cloud security. To detect and neutralise attacks in real time, the proposed IDS uses a variety of algorithms, such as Support Vector Machines (SVM), Random Forest, Decision Trees, K-Nearest Neighbours (KNN), and deep learning techniques like Convolutional Neural Networks (CNN). It minimises false positives while adjusting to new attack vectors. In order to increase response times and detection accuracy, the study combines performance optimisation strategies with feature selection methodologies. It also emphasises how crucial it is to use up-to-date datasets for realistic attack scenario simulation and strong model validation, like CSE-CICIDS2018. Results show that machine learning-based intrusion detection systems (IDS) find novel and sophisticated threats more effectively than traditional signature-based systems. In the end, this study offers insightful information for creating more intelligent and flexible cybersecurity solutions for cloud networks and emphasises the critical function that machine learning plays in protecting sensitive data and guaranteeing the integrity of cloud services across a range of applications.