KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 44 2026 04 10 Machine‑Learning Prediction of Non‑Suicidal Self‑Injury Based on Emotion Dysregulation Facets, Alexithymia, Impulsivity, and Online Social Interaction Patterns Machine‑Learning Prediction of Non‑Suicidal Self‑Injury Based on Emotion Dysregulation Facets, Alexithymia, Impulsivity, and Online Social Interaction Patterns 1 12 10.61838/kman.jayps.5226 EN Journal Article 2025 10 19 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to develop and validate a high-accuracy machine-learning model to predict non-suicidal self-injury (NSSI) by integrating psychological variables (emotion dysregulation, alexithymia, impulsivity) and digital behavior patterns.</p> <p><strong>Methods and Materials:</strong> A cross-sectional study was conducted with one thousand two hundred forty-eight individuals ( ) aged fifteen to twenty-nine. Participants completed validated self-report measures assessing NSSI (Inventory of Statements About Self-Injury), emotion dysregulation (DERS), alexithymia (TAS-20), impulsivity (UPPS-P), and online social interaction patterns. The dataset was partitioned into training ( ) and testing ( ) sets. Three machine-learning models (Support Vector Machines, Random Forest, and XGBoost) were trained and evaluated, with feature importance analyzed using SHapley Additive exPlanations (SHAP).</p> <p><strong>Findings:</strong> The prevalence of lifetime NSSI was ( ). The eXtreme Gradient Boosting (XGBoost) model demonstrated superior predictive performance on the testing set, achieving an accuracy of &nbsp;and an Area Under the Receiver Operating Characteristic Curve (AUC-ROC) of . SHAP analysis identified the most influential predictors of NSSI as the negative emotional valence of online interactions ( ), impulse control difficulties ( ), and negative urgency ( ).</p> <p><strong>Conclusion:</strong> Machine-learning models can accurately predict non-suicidal self-injury, highlighting the critical role of digital emotional distress and impulsivity as primary risk factors.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/jayps/article/download/5226/9496