<![CDATA[The Military Physical Education Center (Pusdikjas) is responsible for providing military physical education to support the duties of the Indonesian Army. Each year, the student admission and education process generates substantial and recurring performance-score data, creating a need for systematic grouping that can support the interpretation of student performance. This research aims to apply the K-Means clustering algorithm to group Pusdikjas education students and to determine the optimal number of clusters using the Silhouette Coefficient. The dataset covers student data from 2018–2022 and contains 13 attributes, with NRP, attitude and behavior, knowledge and skills, and physical scores selected for the clustering process. Data were processed through cleaning, numerical normalization, K-Means clustering, and validation by comparing cluster solutions from K=2 to K=10. The clustering output used 1,427 student records. The results show that K=2 provides the highest Silhouette Coefficient, reaching 0.6543347543126016. Cluster 1 contains 751 students and is interpreted in the manuscript as the very satisfactory category, whereas Cluster 2 contains 676 students and is interpreted as the satisfactory category. The two-cluster solution therefore provides the strongest separation among the tested alternatives. Practically, the resulting grouping can support Pusdikjas in organizing and reviewing student performance data more systematically. Methodologically, the comparison of several K values demonstrates the importance of cluster validation before selecting the final grouping solution.]]>
Copyrights © 2023
