Cloud guard: Optimizing intrusion detection for fortifying privacy defenses with an optimized self adaptive physics-informed neural network

The rapid expansion of cloud computing is fueled by its vast processing capabilities and ability to support data-intensive applications, which highlights the urgent demand for advanced and reliable security solutions. Despite its benefits, the persistent security problems prevent wider use, particularly when it comes to handling sensitive data. To address these, effective intrusion detection is essential for maintaining data confidentiality and integrity. The intricacy of dynamic cloud environments surpasses the capacity of existing methodologies, which underscoring the necessity for advanced solutions. Therefore, a novel approach called Cloud Guard: Optimizing Intrusion Detection for Fortifying Privacy Defenses with an Optimized Self Adaptive Physics-Informed Neural Network (CG-FPD-SAPINN) is proposed in this paper to enhance the intrusion detection in cloud environments. This method leverages Self Adaptive Physics-Informed Neural Network (SAPINN) optimized with the Piranha Search Optimization Algorithm (PFOA). The process begins with data acquisition from the BOT-IoT dataset and proceeds through pre-processing and feature selection phases. Hierarchical Manta Ray Foraging Optimization (HMRFO) is used for optimal feature selection. Then, the SAPINN is used for classification, and its network's weights are optimized by PFOA to improve the efficacy of intrusion detection. The proposed CG-FPD-SAPINN technique is executed in Python using performance metrics, like accuracy, precision, recall, specificity, f-Score, computation time, error rate, latency, throughput, scalability, and resource consumption. The comparative analysis demonstrates that the CG-FPD-SAPINN method outperforms other existing approaches. It exhibits higher recall and Receiver Operating Characteristic (ROC) curve while achieving reduced computation time and latency. The proposed CG-FPD-SAPINN method provides a feasible solution for increasing the safety in cloud computing environments, particularly in intrusion detection. By using advanced techniques, such as SAPINN and PFOA, the method demonstrates better performance in detecting intrusions, thus addressing security concerns and facilitating the secure adoption of cloud technologies.