Advanced Machine Learning Approach for Cyber Attack Detection in Cloud Computing

Cloud computing has become the most important part of modern IT infrastructure, which is scalable and flexible. However, with such wide-scale adoption, new vulnerabilities have been introduced into the system, making it a favorite target for cyber-attacks, especially Distributed Denial-of-Service (DDoS) and Man-in-the-Cloud (MitC) attacks. Security measures in cloud environments are more than ever required to be effective. This study introduces a state-of-the-art approach using machine learning techniques to identify and prevent such attacks; the four machine learning algorithms that have been used here are Support Vector Machine, Artificial Neural Network, Recurrent Neural Network, and k-Nearest Neighbors. A dataset of 3,200 simulated records has been used for the purpose of training and testing; 70% of data were used for training purposes, and 30% of data were used for testing purposes. The models were rated based on precision, running time, and memory usage. The results highlighted that SVM was the strongest model and had an accuracy of 97.86%, while ANN was about 93.4% followed by RNN which was about 89.45% and KNN at 86.7%. This would mean that among the above models, the SVM is the most efficient with respect to real-time attack detection as it is in high precision and has efficiency in resource use. The proposed approach can be very well applied to real-time cloud security systems, offering a robust solution in identifying and mitigating cyber threats in dynamic cloud environments, ensuring enhanced protection against evolving attack vectors.