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...

Full description

Saved in:
Bibliographic Details
Published in2024 2nd International Conference on Cyber Resilience (ICCR) pp. 1 - 5
Main Authors Alrosan, Ayat, Alomoush, Waleed, Youssef, Merna, Nile, Abdelrhman W., Deif, Mohanad A., Gohary, Rania EL
Format Conference Proceeding
LanguageEnglish
Published IEEE 26.02.2024
Subjects
Affective computing
Analytical models
Computational modeling
Human factors
Hyper-parameter tuning
Machine learning algorithms
Medical services
Metaheuristics
Multimodal dataset
Real-time systems
Stress detection
Online AccessGet full text

Cover

More Information
Summary: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.
DOI:10.1109/ICCR61006.2024.10533109