The quick rise in smartphone use has drastically changed how individuals interact with social media, communicate, and obtain information. However, excessive smartphone use has sparked worries about behavioural addiction and its detrimental impacts on social connection, mental health, and productivity. This study suggests evaluating social media usage trends to predict smartphone addiction levels using a machine learning-based method. In addition to general daily, weekly, and monthly usage statistics, the dataset used in this study includes data on the amount of time spent on a variety of apps, including WhatsApp, Instagram, Facebook, Snapchat, LinkedIn, and YouTube. To find important behavioural markers associated with smartphone addiction, data preprocessing and feature analysis are carried out. Next, based on usage patterns, a classification model is built to classify users into various levels of addiction. The results of the experiment show that the suggested model accurately predicts the degree of smartphone addiction, with screen time and social media use appearing as the most significant factors. The results of this study demonstrate how machine learning techniques can be used to identify problematic smartphone usage and promote early intervention tactics for better digital habits.
Ekambaram et al. (Thu,) studied this question.