MACHINE LEARNING MODEL FOR PREDICTION OF SMARTPHONE ADDICTION

Abstract

Smartphone addiction is an increasing concern due to its negative impact on productivity, physical health, and mental well-being. This study developed a machine learning model to predict smartphone addiction using survey data that included demographic information, smartphone usage patterns, and psychological factors such as stress, anxiety, and depression. The data was pre-processed through categorical variable encoding and numerical data normalization to improve model performance. The model was trained on a portion of the dataset and evaluated using accuracy metrics on the remaining data. Results demonstrated high predictive accuracy, with key factors influencing addiction risk including daily phone usage duration, notification-checking frequency, and frequently used app categories. Demographic variables and stress levels also contributed significantly to prediction outcomes. The developed model can assist healthcare professionals in identifying individuals at risk of smartphone addiction and enable early intervention. Additionally, it can help app developers create fewer addictive applications and encourage healthier smartphone usage habits. Overall, the study highlights the effectiveness of machine learning techniques in predicting smartphone addiction, while emphasizing the need for further research using larger and more diverse datasets to improve generalizability and explore broader applications.

Keywords