Observational analysis finds high accuracy in predicting exercise dependence in college students, suggesting strong implications for psychological health monitoring.
Key Points
The stacking ensemble model achieved a mean AUC of 0.96 for predicting exercise dependence risk.
Key predictors include prolonged exercise to gain desired effects and difficulty reducing exercise frequency.
Data was gathered from 2,745 college students using standardized questionnaires assessing various psychological characteristics.
These findings highlight how machine learning methods can be applied effectively to monitor psychological health risks.