<![CDATA[This study compares K-Means Clustering and Hierarchical Clustering to group disaster-prone areas in Papua Barat Province using official data from UPTD Pusdalops BPBD for the period 2020 to November 11, 2025. The dataset covers seven regencies with three aggregated variables: total floods and tidal floods, other disasters, and earthquakes. The research procedure includes data preprocessing, z-score normalization, determining the optimal number of clusters using the Elbow Method and Silhouette Score, and visualizing results with scatter plots and dendrograms. Both algorithms produced two clusters, with Hierarchical Clustering achieving a higher Silhouette Score of 0.4274 compared to 0.339 for K-Means. The scores indicate moderate clustering quality, reflecting the exploratory nature of the study due to the small dataset. Manokwari Regency formed a distinct cluster due to differing disaster characteristics. The novelty lies in employing official BPBD data and Silhouette Score evaluation to compare the two clustering algorithms, supporting data-driven prioritization for disaster mitigation.]]>
Copyrights © 2026
