<![CDATA[This study is motivated by the inequality of primary education quality in Indonesia, reflected in disparities in the number of schools, teachers, students, and facilities across provinces. Data-driven analysis is needed to map these conditions so the government can design more targeted policies. This research applies clustering by comparing K-Means and K-Medoids algorithms using primary school data from the Ministry of Primary and Secondary Education portal. The study follows the CRISP-DM framework, including problem understanding, data preparation, modeling, and evaluation. The optimal cluster number was determined using the Elbow method and Silhouette Score. Results show that K-Means with two clusters achieved the best performance with a Silhouette Score of 0.7069, higher than K-Medoids at 0.6702. The first cluster represents most provinces with smaller education scales, while the second cluster includes larger provinces with significantly more schools, students, and teachers. These findings suggest that K-Means is more suitable for mapping primary education conditions in Indonesia and may support evidence-based policies for educational equity.]]>
Copyrights © 2025
