AI-Based Security Device for Cloud Computing

Cloud computing is a promising and inexpensive solution for scaling and cost-effective computing of high quality. Aspects of cloud computing that are frequently encountered include serious security issues such as confidentiality/integrity threats as well as access control threats. In this paper, we propose an artificial intelligence-based security device which uses machine learning algorithms to detect and mitigate threat in cloud environments in real time. The proposed device is based on a deep learning-based intrusion detection system (IDS) trained on set of cloud traffic datasets including NSL-KDD and CICIDS2017 which uncover anomalies and vulnerabilities to detect and defend against threats. We utilize supervised learning models such as Random Forest and LSTM to obtain highly accurate threat classification and response metrics. Results of experiments show that the proposed device has 96. 3% detection accuracy and low false positives compared to traditional IDS systems. The proposed device can also learn to adapt and response to new threats by continuously learning using continuous learning mechanisms. Our work suggests that intelligent systems can be applied in cloud security frameworks in order to achieve a more resilient and self-sustaining defence architecture. This contribution comes with the benefits of proactive threat management and also improves trust in cloud service providers, especially enterprise in sensitive data management. Further works will explore implementation of federated learning for privacy-preserving model training across distributed cloud systems.