The dual effects of individual and contextual factors on adolescent problematic internet use: Machine learning approaches and SHAP explanations

Key Points

• To identify individual and contextual factors associated with adolescent problematic internet use (PIU) using machine learning methods.
• Applied five machine learning algorithms
• Analyzed interactions between individual and contextual risk factors
• Utilized SHAP explanations for model interpretation
• Risk factors accumulate across systems, impairing adaptability in adolescents
• Mindfulness has cross-system effects that mitigate risks associated with PIU
• Findings suggest the need for targeted interventions