<![CDATA[This study aims to segment QRIS Tap users in public transportation payments through the Bayarind e-wallet application. The K-Means clustering method is employed using factual, non-perceptual data, including users’ age, duration of application usage, frequency of QRIS Tap usage, and average transaction value. The dataset consists of 63 QRIS Tap users in the Jabodetabek area, which has been transformed into numerical form. Cluster evaluation is conducted using the Davies–Bouldin Index (DBI) and Within-Cluster Sum of Squares (WCSS) to determine the optimal number of clusters. The scientific contribution of this study lies in the application of user segmentation based on actual behavioral data (non-perceptual) within the specific context of QRIS Tap usage for public transportation payments, a topic that remains limited in prior studies. Furthermore, this study integrates DBI and WCSS as complementary evaluation metrics to ensure a more objective and robust cluster configuration. The results indicate that a four-cluster configuration (K = 4) provides the most informative segmentation. These clusters represent new users with low activity, loyal users with moderate transaction levels, experienced users with diverse transaction patterns, and premium users with high transaction frequency and value. This segmentation offers empirical insights into QRIS Tap user characteristics and serves as a strategic foundation for decision-making in the development of digital payment systems and public transportation services.]]>
Copyrights © 2026
