<![CDATA[This study aims to compare the performance of two clustering algorithms, K-Means Clustering and K-Medoids Clustering in grouping Indonesian provinces based on forest area by type. The optimal number of clusters was determined using the minimum Davies–Bouldin Index (DBI), while cluster performance was evaluated using the Silhouette Coefficient. Clustering, as one of the key techniques in data mining, automatically classifies data into several groups with similar characteristics. The results reveal differences in the number of clusters produced by the two algorithms. The K-Means method generated four clusters, indicated by its lowest DBI value of 0.515, whereas the K-Medoids method produced three clusters, with a minimum DBI value of 0.559. The clustering performance of K-Means resulted in a Silhouette Coefficient of 0.610, while K-Medoids achieved a higher value of 0.644. Based on these results, the K-Medoids Clustering method with three clusters, demonstrates superior performance in analyzing the grouping of Indonesian provinces by forest area type.]]>
Copyrights © 2025
