Application of Machine Learning in the Field of Medical Science: Sleep Disorders Prediction & Classification

Sleep disorders can profoundly impact an individual's overall health and lifestyle.Conditions such as sleep apnoea and insomnia are often challenging to identify through traditional methods, and expert evaluations can be time-consuming and prone to subjective errors.A highly effective alternative is the application of machine learning algorithms, which can accurately diagnose and categorize sleep disorders.This paper presents an enhanced machine learning approach for detecting sleep disorders, utilizing an open-source dataset from Kaggle to evaluate model effectiveness.The dataset comprises 374 instances and 13 features, reflecting daily activities and sleep habits.Several algorithms were employed and assessed-Support Vector Machines (SVM), Gradient Boosting, Random Forest, K-Nearest Neighbours (KNN), and Logistic Regression-to evaluate their classification performance.The comparative study revealed varying performance levels among the models, with classification accuracies of 88.19%, 89.02%, 88.50%, 89.15%, and 89% for each respective model.Notably, the Random Forest model achieved the highest accuracy at 89.15%.These findings clearly demonstrate that Random Forest outperforms other methods in effectively identifying and classifying different types of sleep disorders.