TY  - EJOU
AU  - Xu, Yajing 
AU  - Xie, Luze 
AU  - Bao, Menghan 
AU  - Yang, Xingyi 
AU  - Chen, Sitong 
AU  - Gao, Zhuoning 

TI  - Exploring Subjective Well-Being in Adolescents: The Role of Mental Health and Addictive Behaviors through Machine Learning
T2  - International Journal of Mental Health Promotion

PY  - 2025
VL  - 27
IS  - 5
SN  - 2049-8543

AB  -  <b>Background:</b> Adolescents’ subjective well-being (SWB) is strongly linked to mental health, academic achievement, social relationships, and quality of life, and is a key predictor of life outcomes in adulthood. Mental health and addictive behaviors are the two main factors influencing SWB. This study aimed to identify key mental health and addictive behavior factors associated with adolescent SWB through machine learning models. <b>Methods:</b> The data for this study comes from the Health Behaviour in School-aged Children (HBSC) survey 2017/18. The study data contains health data from 60,450 adolescents aged 10–16 years. The study used the XGBoost machine learning model to analyze the impact of mental health and addictive behaviors on adolescent SWB. Gain was used to analyze the significance of the variables. The direction of action of the variables and the interaction between the variables were analyzed using the SHapley Additive exPlanations (SHAP) method. <b>Results:</b> The model in this study has an accuracy of 86.7% and an AUC value of 0.85, showing its good predictive performance. Six key variables were filtered through Gain analysis. Feeling low and health as the two most important factors affecting SWB, with these two variables contributing 51.38% and 19.65%, respectively. Friends and thinking body as major factors influencing SWB in mental health. Smoking lifetime and sweets as major factors influencing SWB in addictive behaviors. The interactions and characteristic dependencies between these variables were further analyzed. The results showed that feeling low, friends, and sweets had a positive effect on SWB, while health and smoking lifetime showed a negative effect. In addition, a moderate thinking body contributes to SWB, whereas being too fat and too thin are both associated with decreased levels of SWB. <b>Conclusion:</b> Mental health and addictive behavioral factors such as feeling low, friends, sweets, and smoking lifetime were significant factors influencing SWB. This provides a scientific basis for the development of public health policies and interventions aimed at enhancing adolescent well-being.
KW  - Machine learning; subjective well-being; adolescent; mental health; addictive behavior

DO  - 10.32604/ijmhp.2025.062808