Gaming Addiction Level Prediction Using Random Forest Classification Based on Behavioral and Mental Health Factors

Gaming addiction is a growing concern due to the increasing popularity of online and mobile games. This study aims to predict gaming addiction severity using the Random Forest classification algorithm based on behavioral and mental health factors. The dataset comprises 250 player records with 49 attributes, including gaming duration, screen time, dopamine dependency index, self-control, impulsiveness, stress, and productivity-related factors. After preprocessing, the data were divided into training and testing sets using an 80:20 stratified split. A Random Forest model with 300 trees and balanced class weighting was evaluated using accuracy, precision, recall, F1-score, and a confusion matrix. The model achieved 84% accuracy, with weighted precision, recall, and F1-score of 83%, 84%, and 83%, respectively. However, macro-average precision, recall, and F1-score were 62%, 51%, and 55%, respectively, indicating limited performance on minority classes, particularly the Severe category. Feature importance identified daily playtime hours, dopamine dependency index, total screen time, and maximum consecutive gaming duration as the most important features. The findings indicate good performance on dominant classes but limited minority-class classification, highlighting the need for further data balancing.