Hybrid Optimization-Based Algorithm for Adaptive Threat Detection in Cloud Environments
Venkata Sai Sandeep Velaga · 2025
In the rapidly changing field of cloud computing, traditional security approaches may not be enough to thwart new cyber threats that are now sophisticated and adaptive. This paper proposes a new hybrid optimization-based threat detection model based on adaptive machine learning models and multiple optimization methods, such as Genetic Algorithms and Particle Swarm Optimization (PSO). The proposed model exploits optimization to select relevant features and optimize detection parameters. The hybrid model has the ability to learn adaptive patterns against evolving attacks and dynamically adjust the detection components of the system in real-time, resulting in higher accuracy and fewer false positives. The model comprises four components: (a) a data preprocessing component, (b) an optimized feature selection component, (c) an adaptive detection engine, and (d) a real-time alert and response component. When the proposed threat detection model was evaluated against baseline static models, the results from benchmark cloud security datasets demonstrated the model’s superiority in levels of detection rate, efficiency, and flexibility. This research provides a scalable security model for the challenge of managing threats in dynamic cloud infrastructure, and represents an intelligent proactive and resilient security management solution for analysing new threats.