Analysing the Impact of Lifestyle Factors on Sleep Disorders

Kang, Zhi Huen (2025) Analysing the Impact of Lifestyle Factors on Sleep Disorders. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RMM_Kang Zhi Huen_Full Text.pdf Restricted to Registered users only Download (4MB)

Abstract

The increasing prevalence of sleep disorders such as insomnia and sleep apnea has highlighted the importance of understanding the impact of lifestyle factors on sleep health. This study aims to analyse the relationship between factors like sleep duration, physical activity, stress levels, gender, age, occupation, sleep quality, blood pressure, heart rate and BMI with the occurrence of sleep disorders. A dataset consisting of demographic and lifestyle data was used to predict three categories of sleep disorders: Insomnia, Normal sleep and Sleep Apnea. Machine learning models, including Logistic Regression, Random Forest, Support Vector Machine and CatBoost were employed to classify these disorders. The models were evaluated using metrics such as accuracy, precision, recall, F1-score and AUC to assess their effectiveness in predicting sleep disorders based on lifestyle factors. The findings indicate that CatBoost emerged as the best-performing model with a notable AUC of 0.94 for Sleep Apnea predictions, achieving an accuracy of 72.73%. The study developed a user-friendly GUI where users input their data, and the model provides a prediction with a confidence level, visually displaying Normal sleep in green and Non-Normal disorders in red for easy understanding. In conclusion, the CatBoost model proves to be an effective tool for predicting sleep disorders based on lifestyle and health data. Future improvements could include adding more features, conducting further hyperparameter tuning, and real-world testing to help individuals manage their sleep health through early detection and personalised guidance. Keywords: Sleep health, Sleep disorder, Insomnia, Sleep apnea, CatBoost, Machine learning, Health prediction.