Enhancing Portal System Resilience with a Modified Lion Optimization Algorithm (MLOA) for Cyber Threat Detection

Abstract

This research presents a novel cyber threat detection framework that integrates the Modified Lion Optimization Algorithm (MLOA) with a one-class classification approach to improve the resilience of portal systems against denial-of-service attacks, Man-in-the-Middle attacks, and data breaches. The proposed model enhances anomaly detection by optimizing decision boundaries in high-dimensional datasets, leveraging adaptive threshold tuning, dynamic feature selection, and real-time monitoring. Experimental evaluations demonstrate that the MLOA-based detection model significantly outperforms traditional clustering-based methods across varying levels of attack complexity. It achieves a recall of 0.97, accuracy of 0.98, precision of 0.96, and an area under the receiver operating characteristic curve (ROC-AUC) score of 0.97 for simple anomalies, while maintaining strong performance for moderate and complex anomalies, with recall values of 0.92 and 0.90 and ROC-AUC scores of 0.94 and 0.92. These findings validate the effectiveness of the proposed approach in detecting zero-day attacks and evolving cybersecurity threats, offering a scalable, high-performance anomaly detection solution for modern portal systems. This study further establishes the practical application of nature-inspired optimization algorithms in cybersecurity, reinforcing the importance of AI-driven threat detection in protecting digital infrastructure.