Adaptive Quantum Particle Swarm Optimization with Deep Ensemble Learning for Smart Health Monitoring in Big Data Systems

Abstract

In the realm of smart health monitoring within big data systems, traditional machine learning algorithms face challenges related to accuracy and computational efficiency. Addressing this research gap, we propose the Deep Ensemble Learning (DEL) method, specifically leveraging Adaptive Quantum Particle Swarm Optimization (AQPSO-DEL). This method aims to enhance predictive performance by integrating quantum mechanics principles with deep ensemble learning frameworks. Using two comprehensive datasets, the proposed AQPSO-DEL method was evaluated against established algorithms such as JA-DEL, SSA-DEL, COA-DEL, DOA-DEL, and HDCO-DEL. Our results demonstrate that AQPSO-DEL consistently achieves the highest accuracy (97.45%), sensitivity (97.78%), and specificity (98.22%), outperforming its counterparts significantly. Notably, AQPSO-DEL also recorded superior precision, minimized false positive and negative rates, and an exceptional F1-score of 95.34%. These improvements underscore the robustness and reliability of AQPSO-DEL, offering a marked performance enhancement of up to 5.78% over existing methods. This work highlights the transformative potential of combining deep ensemble learning with advanced quantum optimization techniques to deliver highly accurate and efficient smart health monitoring solutions.