An Intelligent Data Leakage Prevention in Cloud Computing using Golden Jackal Optimization

This research proposes an intelligent data leakage prevention framework for cloud computing environments using the Golden Jackal Optimization (GJO) algorithm, designed to enhance detection accuracy and minimize false alarms in real-time. By optimizing both feature selection and classifier parameters, the GJO-based model significantly improves performance across key metrics, including Accuracy, True Positive Rate (TPR), False Positive Rate (FPR), and Recall. Evaluated on the UNSW-NB15 dataset and simulated using the SUMO environment to replicate realistic cloud traffic, the proposed approach outperforms traditional methods such as ClamAV, Suricata, and XGBoost. Experimental results demonstrate that the GJO model achieves higher detection rates while reducing false positives, offering up to 15–25% improvement in precision-related metrics. Its scalability and adaptability make it highly suitable for deployment in dynamic and large-scale cloud infrastructures. This work highlights the potential of nature-inspired optimization algorithms in building intelligent, efficient, and resilient cloud security systems capable of mitigating advanced data leakage threats.