Enhancing Security in Cloud Computing and Protocols Using Harris Hawks Optimizer with Deep Learning for Intrusion Detection
Rajeev Kumar Tiwari, S. Murugappan · 2023
The development of cloud computing (CC) has revolutionized the approach to information technology (IT) services that can managed and delivered, with flexibility, providing unprecedented scalability, and cost efficiency. As organizations gradually depend on cloud-based structure, the possible attack surface widens, requiring robust intrusion detection systems (IDS) for safeguarding sensitive data and services. By leveraging developed machine learning (ML) approaches and anomaly detection systems, cloud-based IDS proactively recognizes and responds to security attacks, and data breaches, protecting against unauthorized access, and other malicious actions. Therefore, this study presents a novel Harris Hawks Optimizer with Deep Learning assisted Intrusion Detection Approach (HHODL-IDA) for CC environment. The HHODL-IDA technique improves security level in the CC environment via automated intrusion detection process. Primarily, the HHODL-IDA technique applies min-max normalization which ensures uniform data preprocessing. Besides, the HHODL-IDA technique uses stacked autoencoders (SAE), which derive intricate features and enable accurate detection of anomalies. Eventually, the HHO method can be deployed for optimizing the hyperparameters of the SAE method. The experimental assessment of the HHODL-IDA approach can be investigated on benchmark IDS datasets. The extensive outcomes infer that the HHODL-IDA approach gains enhanced intrusion detection performance over other recent models in the CC environment.