<![CDATA[Social media has become an integral part of students’ lives, yet excessive use often leads to symptoms of addiction that negatively affect mental health and academic performance. This study aims to cluster the levels of social media addiction among students and university students using the K-Means Clustering algorithm as an unsupervised learning approach. The dataset was obtained from the Kaggle platform, containing variables such as daily usage duration, access frequency, sleep disturbance, and psychological impact. The Elbow Method was employed to determine the optimal number of clusters, while Principal Component Analysis (PCA) was used for visualization. The results grouped respondents into three categories: mild addiction (46.8%), moderate addiction (22.6%), and severe addiction (30.6%). A strong correlation was observed between high access frequency and symptoms such as sleep disruption and decreased concentration. These findings highlight the importance of designing data-driven prevention strategies within educational environments and provide a foundation for further institutional interventions to maintain digital balance among youth.]]>
Copyrights © 2026
