Understanding Digital Behavior and Smartphone Addiction Through Machine Learning: Implications for Social Life in the Digital Era

Smartphone usage has become an integral part of daily life in the digital era, but excessive use may lead to smartphone addiction and influence individual behavior in digital society. This study aims to analyze digital behavior patterns associated with smartphone addiction using machine learning techniques. The dataset used in this research consists of 7,500 smartphone users with variables describing demographic characteristics and digital activity patterns, including screen time, social media usage, gaming activity, notifications, and application usage frequency. Several machine learning algorithms were implemented to classify smartphone addiction, including Logistic Regression, Random Forest, and Support Vector Machine. The results show that the Random Forest model achieved the best performance with an accuracy of 93.4 percent and an F1 score of 0.953, indicating strong predictive capability in identifying addicted users. Feature importance analysis reveals that social media usage hours, daily screen time, and weekend screen time are the most influential predictors of smartphone addiction. In addition, clustering analysis identified two distinct behavioral groups of smartphone users characterized by moderate and intensive digital engagement patterns. These findings highlight the significant role of digital behavior patterns in shaping smartphone addiction and demonstrate that machine learning approaches can effectively identify behavioral indicators associated with excessive smartphone use. The results provide insights into understanding smartphone addiction within the context of digital society and contribute to the development of strategies aimed at promoting healthier digital behavior.