Earthquakes are natural disasters that cannot be prevented or avoided. One of the areas affected is the West Sumatra region, where West Sumatra is one of the regions in Indonesia which is in the Sumatra basin which is vulnerable to earthquakes. Therefore, density-based clustering analysis can be carried out which aims to produce a point map of earthquake-prone areas in the West Sumatra region using the Density Based Spatial Clustering of Application with Noise (DBSCAN) method. In implementing the DBSCAN algorithm, epsilon and minpts parameters are required using the K-Nearest Neighbors method with evaluation of results using the Silhouette Coefficient. The results of DBSCAN clustering using KNN input parameters obtained a total of 3 clusters and 1 noise with a silhouette coefficient value of 0.310 from the 2010-2023 data period. However, from the testing stage without using KNN, we got a high silhouette score, namely 0.890 with 2 clusters and 1 noise.
Copyrights © 2024