<![CDATA[Indonesia is one of the countries with a high vulnerability to earthquakes due to its location at the convergence of three major tectonic plates: the Indo-Australian, Eurasian, and Pacific plates. As a result of this interaction, seismic activity is highly frequent across various regions. Understanding the distribution patterns of earthquakes is essential for disaster risk mitigation. One approach used to analyze these patterns is clustering, particularly using the DBSCAN and Meanshift algorithms, which can group spatial data without predefining the number of clusters. This study aims to compare the effectiveness of both algorithms in clustering earthquake data based on spatial parameters, namely latitude and longitude. Evaluation was conducted using cluster visualization and the Silhouette Score as the clustering validity metric. The results show that DBSCAN produces more optimal clustering with a Silhouette Score of 0.930028, higher than Meanshift's score of 0.90103. DBSCAN is also capable of detecting relevant outliers in earthquake analysis, while Meanshift generates more clusters but with less separation. Using spatial parameters such as latitude and longitude, DBSCAN is considered more effective in identifying the spatial distribution patterns of seismic activity in Indonesia based on earthquake data. This research supports the development of decision support systems for earthquake disaster mitigation and serves as a reference for selecting appropriate clustering methods for spatial data analysis.]]>
