<![CDATA[Indonesia is one of the regions with a high level of natural disaster risk because it is geographically located between three major tectonic plates of the world. This condition makes Indonesia vulnerable to various types of disasters that occur almost every year in different regions, resulting in material losses and casualties. This study aims to compare the performance of two clustering algorithms, namely K-Means and K-Medoids, in grouping provinces in Indonesia based on their level of vulnerability to natural disasters, evaluated using the Davies–Bouldin Index (DBI). The analysis results show that the K-Means algorithm provides the best performance with a DBI value of 0.3932 at K = 3, forming three main categories: provinces with low to moderate vulnerability (35 provinces), high vulnerability (2 provinces), and very high vulnerability (1 province). Meanwhile, the K-Medoids algorithm produces the lowest DBI value of 0.5860 at K = 2, which divides provinces into two major groups of disaster risk levels. Based on this comparison, K-Means is considered more effective for mapping disaster vulnerability levels in Indonesia because it is able to represent risk patterns in greater detail. These findings are expected to serve as a reference for relevant agencies in formulating mitigation strategies and determining priority handling for regions with high disaster risk.]]>
Copyrights © 2026
