<![CDATA[Indonesia is one of the countries with the highest levels of seismic activity in the world because it is located at the meeting point of three major plates. The high potential for earthquakes requires a data-based approach to map vulnerable areas more accurately. This study aims to group earthquake-prone areas in Indonesia using the K-Means and DBSCAN clustering algorithms. The dataset used includes spatial data (latitude, longitude) and seismic data (magnitude, depth, phasecount, azimuth_gap) obtained from the BMKG earthquake catalog for the period 2008–2025. The study begins with the data preprocessing stage, which includes data cleaning, type conversion, feature selection, missing value imputation, outlier detection and removal, and normalization. Furthermore, the clustering algorithm is applied in three main scenarios, namely spatial data, seismic data, and a combination of spatial and seismic data. Evaluation using Silhouette Score and Davies-Bouldin Index (DBI) metrics shows that the K-Means algorithm provides better cluster separation, with a DBI value of 1.2551 in the combined scenario, while the DBSCAN algorithm tends to form only one dominant cluster and is sensitive to the presence of outliers. The final result of this study produces a map of earthquake-prone areas in Indonesia, which are divided into several clusters with different risk characteristics. The cluster with the highest concentration includes areas such as western Sumatra, the southern coast of Java, and parts of Maluku and Papua, which have historically been recorded as having higher earthquake frequency and magnitude. Meanwhile, other clusters covering areas such as Kalimantan and parts of Sulawesi show lower seismic activity intensity.]]>
Copyrights © 2025
