KMAN Publication Inc. (KMANPUB) Journal of Assessment and Research in Applied Counseling (JARAC) 8 Serial Number 30 2026 07 01 CatBoost Classification of Digital Burnout Based on Technostress, Online Boundary Control, and Sleep Procrastination CatBoost Classification of Digital Burnout Based on Technostress, Online Boundary Control, and Sleep Procrastination 1 15 EN Journal Article 2026 02 27 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to classify digital burnout severity among adult digital technology users in Canada using the CatBoost machine learning algorithm based on technostress, online boundary control, and sleep procrastination.</p> <p><strong>Methods and Materials:</strong> This applied, quantitative, cross-sectional, and predictive classification study was conducted among 486 adult digital technology users residing in Canada. Participants were selected through convenience sampling and completed standardized self-report measures assessing digital burnout, technostress, online boundary control, and sleep procrastination. Digital burnout was categorized into low, moderate, and high classes and used as the target outcome variable. Technostress, online boundary control, and sleep procrastination were entered as predictor variables. The dataset was divided into training and testing subsets, and the CatBoost classifier was developed using cross-validation and hyperparameter tuning. Model performance was evaluated using accuracy, precision, recall, F1-score, ROC-AUC, Cohen’s kappa, confusion matrix, and feature importance indices.</p> <p><strong>Findings:</strong> The CatBoost model demonstrated strong classification performance in predicting digital burnout severity. The overall accuracy of the model was 0.84, with a weighted F1-score of 0.84 and a macro-average ROC-AUC of 0.91. Cohen’s kappa coefficient was 0.75, indicating substantial agreement between observed and predicted burnout classes. Class-specific results showed that the model classified low digital burnout with precision = 0.88, recall = 0.91, and F1-score = 0.89; moderate digital burnout with precision = 0.79, recall = 0.79, and F1-score = 0.79; and high digital burnout with precision = 0.84, recall = 0.81, and F1-score = 0.82. Feature importance analysis showed that technostress was the strongest predictor, followed by sleep procrastination and online boundary control.</p> <p><strong>Conclusion:</strong> The findings indicate that CatBoost can accurately classify digital burnout severity based on psychological and behavioral predictors. Technostress, sleep procrastination, and online boundary control appear to be meaningful indicators for identifying individuals at risk of higher digital burnout.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/jarac/article/download/5809/10995