Paper Key : IRJ************287
Author: B.hasini,G.s.guna Varshini,A.poojitha,C.thulasi,Mr.n.vijaya Kumar,Dr.r.karunia Krishnapriya,Mr.pandreti Praveen,Mr.v Shaik Mohammad Shahil
Date Published: 01 Apr 2025
Abstract
There are serious hazards to ones health and general well-being from sleep problems, including insomnia and sleep apnea. Polysomnography (PSG) and other traditional diagnostic techniques are costly, time-consumingand interpretatively complex. The usefulness of ensemble learning methods, specifically stacking and boosting, for automated sleep problem identification is investigated in this work. We apply and evaluate the stacking classifier, XG Boost, AdaBoost, and gradient boosting models using the Sleep Health and Lifestyle Dataset. Furthermore, in order to improve classification accuracy, we suggest a novel stacking technique that incorporates random forest, gradient boosting, and artificial neural networks (ANNs). The models are assessed using performance criteria such as F1-score, recall, accuracy, and precision. According to experimental results, the hybrid stacking model that incorporates ANNs performs better than classic ensemble approaches by identifying intricate patterns in the data, while boosting techniques enhance robustness and generalization.Our study offers tailored insights and suggestions to assist people in managing and overcoming their diagnosed sleep disorders, going beyond simple diagnosis. These suggestions are based on possible medical interventions, lifestyle changes, and sleep hygiene techniques that were identified from the dataset trends. In addition to providing practical advice for enhancing sleep health, the results demonstrate the promise of ensemble learning in conjunction with deep learning as a scalable and economical method of diagnosing sleep disorders. Keywords: Ensemble Learning, Stacking Classifier, Boosting Techniques, Random Forest, Gradient Boosting, Artificial Neural Networks (ANNs), Machine Learning in Healthcare, Sleep Health and Lifestyle Dataset, Polysomnography (PSG) Alternative, Predictive Analytics for sleep Disorders, XGBoost and AdaBoost, Sleep Hygiene Recommendations
DOI Requested