AI-Driven Digital Well-being: Developing Machine Learning Model to Predict and Mitigate Internet Addiction

Background: Internet addiction has become a major public health issue due to the increased dependence on digital technology, affecting mental health and overall well-being. Artificial intelligence (AI) offers innovative approaches to predicting and mitigating excessive internet use. Objective: This study aims to develop and evaluate AI-driven machine learning models for predicting and mitigating internet addiction by analyzing behavioral patterns and psychological indicators. Methods: Open-access datasets from “Kaggle”, such as “Smartphone Usage Data” and “Social Media Usage and Mental Health”, were analyzed using machine learning and deep learning models, including Random Forest, XGBoost, Neural Networks, and Natural Language Processing (NLP) techniques. Model performance was assessed based on accuracy, precision, recall, F1-score, and AUC-ROC. Results: Neural Networks and XGBoost achieved the highest accuracy (91% and 90%, respectively), surpassing traditional models like Logistic Regression and SVM. Clustering and anomaly detection techniques provided further insights into user behavior, while NLP revealed emotional and thematic patterns associated with addiction. Conclusion: AI-driven models effectively predict and classify internet addiction, offering scalable and personalized interventions to promote digital well-being. Future research should focus on addressing ethical concerns and improving real-time deployment of these models.