Classification Of Sleep Disorder Using Advanced Machine Learning Technology

Abstract

Sleep disorders, particularly insomnia and sleep apnea, significantly affect physical health, cognitive performance, and overall quality of life, making timely and accurate diagnosis essential. Conventional manual assessment of sleep-related conditions is time-consuming, subjective, and prone to diagnostic inconsistencies, motivating the adoption of intelligent classification techniques. A publicly available Sleep Health and Lifestyle dataset containing 400 instances with 13 attributes representing demographic, physiological, and lifestyle characteristics was utilized for analysis. Data preprocessing included duplicate removal, missing-value handling, feature extraction, label encoding, feature selection, class balancing using SMOTE, feature scaling with Min-Max normalization, and train–test splitting to improve model reliability. Multiple machine learning models, including KNearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Artificial Neural Network (ANN), Gaussian Naïve Bayes, Gradient Boosting, and a hybrid Voting Classifier integrating Decision Tree, Bagging, and Random Forest, were implemented and compared. Performance was assessed using accuracy, precision, recall, F1- score, confusion matrix, and classification report. The optimized Decision Tree achieved an accuracy of 90.7%, while the hybrid Voting Classifier delivered the highest classification accuracy of 97.3%, demonstrating superior predictive capability and robustness. The proposed intelligent classification framework enhances diagnostic performance and provides a practical foundation for efficient web-based sleep disorder assessment. Keywords— Sleep Disorder Classification, Machine Learning, Ensemble Learning, Voting Classifier, Sleep Health and Lifestyle Dataset, Healthcare Analytics.