IMPLEMENTATION OF THE DBSCAN METHOD FOR CLUSTER MAPPING OF EARTHQUAKE SPREAD LOCATION

  • Bariklana M
  • Fauzan A
N/ACitations
Citations of this article
22Readers
Mendeley users who have this article in their library.

Abstract

West Java area is located on the Pacific Circum and Mediterranean Circum routes, this causes West Java area to be an unstable area that is characterized by many active working volcanoes and frequent earthquakes. An analysis of the grouping of earthquake data in West Java Province area is urgently needed. The purpose of this study was to classify areas based on the density of earthquake occurrence areas in West Java using Density-Based Spatial Clustering of Application with Noise (DBSCAN). The population in this study are all earthquake events occurred in 2021. While the sample used in this study is data on the location of the distribution of earthquakes in West Java Province in 2021 taken from the BMKG online data website at dataonline.bmkg.go.id. This research began with nearest-neighbor analysis to see patterns of data distribution. If the data distribution pattern is grouped, then DBSCAN analysis can be continued. The DBSCAN algorithm uses a combination of parameters, namely minimum points (MinPts) and epsilon (Eps). Cluster results are evaluated using the silhouette coefficient. Then, in this study, deeper data exploration was carried out in three ways, namely: (1) Clustering based on the highest silhouette value, (2) clustering by lowering the MinPts value, and (3) clustering based on the smallest upper limit (supremum) value of the silhouette coefficient. The data exploration here aimed to form more clusters while still considering the silhouette coefficient value limits so that there are more areas prone to earthquakes but also maintaining the validity of the results obtained. Next, determine the best cluster results by comparing the cluster results obtained. The best cluster results were obtained at Eps=10000 and MinPts=3 which formed 12 clusters with a silhouette coefficient value of 0.713, which means that the clusters have a strong structure. It is hoped that the information regarding the grouping of areas where earthquakes frequently occur can be used as a form of earthquake disaster mitigation and minimize the impact of losses due to the earthquake.

Cite

CITATION STYLE

APA

Bariklana, M., & Fauzan, A. (2023). IMPLEMENTATION OF THE DBSCAN METHOD FOR CLUSTER MAPPING OF EARTHQUAKE SPREAD LOCATION. BAREKENG: Jurnal Ilmu Matematika Dan Terapan, 17(2), 0867–0878. https://doi.org/10.30598/barekengvol17iss2pp0867-0878

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free