Comparative Analysis of Machine Learning Methods for Enhancing Sleep Efficiency and Prediction

Sleep efficiency corresponds to the fundamental aspect of human wellbeing in the context of improving health, productivity, education, and overall quality of life (QoL) for individuals and the community. This study presents a comprehensive analysis of sleep efficiency by evaluating various influential factors. We analyze the impact on sleep efficiency by considering age, gender, sleep duration, percentages of Rapid Eye Movement (REM) sleep, deep sleep, light sleep, awakenings, caffeine consumption, alcohol consumption, smoking status, and exercise frequency. The findings highlight the potential of predicting sleep efficiency and offer valuable insights. Many Unseen patterns emerged from the analysis, e.g., women in their 50s and men in their 60s exhibited increased sleep efficiency. Surprisingly, caffeine consumption did not significantly affect sleep efficiency, while higher alcohol consumption and smoking status were correlated with lower efficiency. Exercise frequency shows a slight positive correlation with sleep efficiency. This paper uses machine learning algorithms, including Linear Regression, Decision Tree, Random Forest, and Gradient Boosting Regressor to predict sleep efficiency. Among these, the Random Forest model outperformed the others, demonstrating the highest sleep efficiency prediction accuracy based on the factors considered.