Prediction of smartphone overdependence and analysis of its influencing factors among older adults based on machine learning.
This study used panel data from South Korea's 2023 Smartphone Overdependence Survey to build and compare machine-learning models predicting smartphone overdependence among adults aged 60+. Among evaluated classifiers, XGBoost had the best reported predictive performance (accuracy 0.925). The most important predictors were demographics, time composition of smartphone use, awareness of overdependence, and content of smartphone use.
Key points
- The study focuses on smartphone overdependence among South Korean adults aged 60 and above using survey panel data.
- Multiple binary classifiers were evaluated, including XGBoost, SVM, logistic regression, KNN, decision tree, and naive Bayes.
- Model performance was assessed using metrics such as accuracy, precision, recall, F1 score, and AUC.
- XGBoost was reported as the best-performing model with an accuracy of 0.925.
- Feature importance suggested demographics, time composition of use, awareness of overdependence, and content of use as key predictors.
- The paper frames machine learning as useful for identifying risk and influencing factors for smartphone overdependence.
Referenced studies & papers
Relevant papers in OpenMel
Source:
Open original
AI-generated summaries may be incomplete or incorrect. This content is for informational purposes only and is not medical advice.
AI-generated summaries may be incomplete or incorrect. This content is for informational purposes only and is not medical advice.
Comments
Log in to comment.
No comments yet.