Behavioral Pattern Analysis for Smartphone Addiction Prediction Using Machine Learning

This study presents a machine learning framework for predicting smartphone addiction using behavioral usage data from 7,500 users. Features include screen time, social media usage, gaming activity, sleep patterns, and notifications. The pipeline incorporates feature engineering (Social Dominance Ratio, Screen-Sleep Ratio) and 5-fold stratified cross-validation. Random Forest achieved the best performance with an F1-score of 0.9554, outperforming Logistic Regression and Gradient Boosting. Feature importance analysis shows social media usage and screen time as dominant predictors. A behavioral transition zone (2.0–2.5 hours of social media use) was identified, indicating addiction as a gradient phenomenon. The work contributes a reproducible ML pipeline and insights for early detection systems in digital health.