Hyper-Parameters Tuning Using Meta-Heuristic Algorithms for Nurses Stress Detection

This research utilizes machine learning techniques, particularly Random Forest, to examine the impact of mental stress on nurses. Stress, arising from challenging events or demanding conditions (stressors), poses significant risks to mental and physical well-being. Real-time data collection through wearable devices includes physiological measurements, behavioral patterns, and environmental factors, creating a comprehensive method for stress monitoring. Personalized and adaptive models, crucial for accurate stress detection due to its subjective nature, are explored. The study integrates data science and machine learning to enhance efficiency in various stress detection methods, employing meta-heuristics search algorithms. Comparative analysis yielded promising results with a 96.63% accuracy and a 95.76% f1 score.