Predictive model based on machine learning to identify sleep-related health problems

Abstract

Sleep quality has become a growing public health issue worldwide, mainly due to a lack of awareness about its long-term consequences. Despite existing strategies to address this problem, there remains a need for more effective approaches. In this study, an early detection model for sleep disorders was implemented using the extreme gradient boosting (XGBoost) algorithm, following the knowledge discovery in databases (KDD) methodology, which includes the phases of selection, preprocessing, transformation, data mining, and interpretation. A dataset extracted from the Kaggle platform in CSV format was used, consisting of 374 records. With an overall accuracy of 91.5%, a recall of 100% for the insomnia class, and a precision of 100% for sleep apnea, the proposed model demonstrated exceptional performance. It also received an area under the curve (AUC) of 0.909 and an average F1-score of 0.913. With a mean accuracy of 91%, a 95% confidence interval (0.89–0.94), and a p-value of 0.0012, cross-validation confirmed its robustness and showed a statistically significant change from the baseline model. The error rates remained within clinically acceptable ranges, confirming its applicability as a diagnostic support tool. Overall, the results demonstrate the effectiveness of the model in identifying patterns related to sleep disorders.