Applying Machine Learning Algorithms for the Classification of Sleep Disorders Accuracy and Automating Diagnosis with the Sleep Health Dataset

Sleep disorders affect a significant portion of the population and are associated with various health risks, including cardiovascular disease and psychological distress. This study applies machine learning algorithms to classify sleep disorders based on the Sleep Health dataset. Several models, including Stacking Classifier and Voting Classifier, were developed to automate diagnosis. The data was pre-processed through encoding and feature engineering, followed by model training using Random Forest, Support Vector Classifier, and XGBoost within an ensemble learning framework. The Stacking Classifier achieved an accuracy of 92%, outperforming the Voting Classifier, which also achieved 92% accuracy. Both models were evaluated through crossvalidation, and the Stacking Classifier demonstrated superior performance. The results suggest that automated classification of sleep disorders through machine learning models can aid in early diagnosis, providing an efficient tool for healthcare professionals.