Clustering Age of Internet User using K-Means Algorithm

Abstract

This paper applied the K-Means clustering algorithm to analyze age-based segmentation of internet users and to identify dominant behavioral patterns across demographic groups. We utilized aggregated datasets compiled from multiple academic journal studies and implemented data normalization to ensure balanced representation across heterogeneous samples. We used the Elbow Method and Within-Cluster Sum of Squares (WCSS) evaluation to determine the optimal number of clusters. The analysis identified three stable clusters representing Early Adopters (≤25 years), Productive Users (26–45 years), and Late Majority (>45 years). The results showed that younger users demonstrated high-frequency, entertainment-oriented, and adaptive digital behavior. Middle-aged users exhibited structured, productivity-driven engagement characterized by digital transactions, professional collaboration, and e-learning activities. Older users displayed more selective internet usage, primarily focused on communication and health-related information. The clustering outcomes aligned with findings reported in prior journal studies, which strengthened the validity and interpretability of the segmentation model. This study confirms that age significantly influences digital behavior patterns and demonstrates that K-Means clustering provides an effective approach for demographic segmentation to support strategic digital service development.