Implementation of DBSCAN for Earthquake Clustering in Indonesia with Potential Surface Damage

Indonesia is one of the countries with the highest seismic activity in the world due to its location at the convergence of three major tectonic plates. Understanding earthquake distribution patterns is crucial for disaster mitigation efforts and policy planning. This study applies the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to cluster earthquake data in Indonesia based on magnitude and depth. The data used is secondary data from the Meteorology, Climatology, and Geophysics Agency (BMKG) for the period of January–December 2023. The research stages include data collection and preprocessing, applying the DBSCAN algorithm with the selection of Eps and MinPts parameters, and evaluating the clustering results using the silhouette coefficient and Davies-Bouldin Index (DBI). The results show that the combination of Eps = 0.5 and MinPts = 5 produces clusters with a silhouette coefficient of 0.3959 and a DBI of 0.7384, indicating a fairly good cluster structure. Visualization results reveal high-density clusters in active seismic zones and several smaller clusters representing specific earthquake characteristics. This study provides insights into the earthquake distribution patterns in Indonesia and demonstrates that DBSCAN effectively identifies complex cluster structures. The findings can serve as a reference for seismological studies and support earthquake disaster mitigation efforts.