<![CDATA[The increasing use of artificial intelligence and digital platforms among students has created new opportunities for learning support, academic assistance, and digital interaction. However, intensive platform usage may also be associated with mental health concerns, sleep disruption, and negative effects on students’ daily life. This study aims to develop and evaluate machine learning models for predicting the overall impact of AI and digital platform usage among students by integrating demographic, behavioral, sleep-related, and mental health-related variables. The dataset consisted of 1,705 student records with features including age, gender, academiclevel, country, average daily usage hours, most-used platform, sleep hours per night, and mental health score. The target variable was Overall_Impact, categorized into Negative, Neutral, and Positive classes. Six supervised machine learning algorithms were evaluated: Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, K-Nearest Neighbors, and Gradient Boosting. Model performance was assessed using accuracy, precision, recall, F1-score, Cohen’s Kappa, MAE, RMSE, ROC-AUC, and confusion matrix. The results showed that Random Forest achieved the best performance, with an accuracy of 99.71%, F1-macro of 99.52%, Cohen’s Kappa of 0.9950, and ROC-AUC of 0.9994 on the testing set. Feature importance analysis revealed that Mental_Health_Score, Sleep_Hours_Per_Night, and Avg_Daily_Usage_Hours were the most influential predictors. The findings indicate that machine learning can effectively predict the impact of digital platform usage and provide useful insights for AI-driven health informatics and student well-being monitoring. However, further validation using longitudinal and clinically grounded datasets is recommended.]]>
Copyrights © 2026
