KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 45 2026 05 01 Support Vector Machine Prediction of Adolescent Anxiety Based on Nomophobia, Fear of Missing Out (FoMO), Perfectionism, Cognitive Fusion, and Parent–Child Communication Support Vector Machine Prediction of Adolescent Anxiety Based on Nomophobia, Fear of Missing Out (FoMO), Perfectionism, Cognitive Fusion, and Parent–Child Communication 1 13 10.61838/kman.jayps.5420 EN Journal Article 2025 09 15 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> The present study aimed to predict adolescent anxiety using a Support Vector Machine (SVM) model based on nomophobia, Fear of Missing Out (FoMO), perfectionism, cognitive fusion, and parent–child communication among Tunisian adolescents.</p> <p><strong>Methods and Materials: </strong>The study employed a cross-sectional correlational design with a machine learning approach. The statistical population consisted of secondary school adolescents in Tunisia during the 2025–2026 academic year. Using multistage cluster random sampling, 742 students were selected, of whom 718 valid questionnaires were retained for final analysis after data screening. Data were collected using the Beck Anxiety Inventory, Nomophobia Questionnaire, Fear of Missing Out Scale, Multidimensional Perfectionism Scale, Cognitive Fusion Questionnaire, and Parent–Adolescent Communication Scale. Descriptive statistics and Pearson correlation analyses were conducted using SPSS-27, while machine learning analyses were performed using Python and Scikit-learn libraries. The dataset was divided into training and testing subsets using an 80/20 ratio, and a radial basis function kernel Support Vector Machine model with five-fold cross-validation was implemented to predict adolescent anxiety.</p> <p><strong>Findings: </strong>The findings indicated significant positive correlations between adolescent anxiety and nomophobia, FoMO, perfectionism, and cognitive fusion, while parent–child communication demonstrated a significant negative relationship with anxiety symptoms. Cognitive fusion emerged as the strongest predictor variable, followed by nomophobia, parent–child communication, FoMO, and perfectionism. The Support Vector Machine model demonstrated high predictive performance, achieving 88.76% classification accuracy in the testing dataset, with an F1-score of 87.16% and an area under the curve (AUC) value of 0.91. The confusion matrix additionally revealed high sensitivity and specificity in distinguishing adolescents with elevated anxiety symptoms from those with lower anxiety levels.</p> <p><strong>Conclusion: </strong>The results demonstrated that adolescent anxiety can be effectively predicted through the combined influence of cognitive, technological, emotional, and family-related factors using machine learning techniques. Cognitive fusion and nomophobia represented the strongest psychological risk factors, while healthy parent–child communication functioned as a protective factor against anxiety symptoms. The findings support the usefulness of Support Vector Machine models in adolescent mental health screening and highlight the importance of addressing maladaptive cognitive processes, digital dependency, and family communication patterns in preventive and intervention programs targeting adolescent anxiety.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/jayps/article/download/5420/9861