Implementation of ST-DBSCAN Algorithm to Cluster Regency / City in West Java Province Based on Natural Disaster Occurrence in 2020-2022

Zahrah Nur Fauziyah

doi:10.21776/ub.jasds.2025.002.01.4

Authors

Zahrah Nur Fauziyah Department of Statistics, Universitas Brawijaya, Indonesia

DOI:

https://doi.org/10.21776/ub.jasds.2025.002.01.4

Keywords:

Cluster Analysis, Natural Disasters, West Java, ST-DBSCAN

Abstract

Natural disaster is an occurrence that threatens and disrupts the lives of living things caused by natural factors and causes environmental damage, casualties, and so on. One of the regions in Indonesia that experienced natural disasters with high intensity in the 2020-2022 timeframe is West Java Province. The purpose of this study is to group regency/city in West Java Province based on the occurrence of natural disasters in 2020-2022 and the characteristics of each cluster formed. This research uses spatio temporal data applied to the Spatio Temporal-Density Based Clustering of Application with Noise (ST-DBSCAN) algorithm using 3 variables, namely the incidence of floods, landslides, and tornadoes in West Java Province in 2020-2022. This study uses the parameters epsilon 1 of 10, epsilon 2 of 30, and minimum points of 3 using the K-NN algorithm to form clusters from the ST-DBSCAN algorithm. The results showed that 2 clusters and 11 noise points were formed with a silhouette coefficient value of 0.5176, where cluster 1 consisted of 66 regency/city, cluster 2 consisted of 4 regency/city and 11 regency/city were included in the noise.

References

Birant, D., & Kut, A. (2007). ST-DBSCAN: An Algorithm for Clustering Spatial – Temporal Data. Data & Knowledge Engineering, 60.

Han, J., Kambe, M., & Pei, J. (2011). Data Mining: Concepts and Techniques. In Data Mining: Concepts and Techniques. https://doi.org/10.1016/C2009-0-61819-5

Purwanto, U. Y. (2012). Penggerombolan Spasial Hotspot Kebakaran Hutan dan Lahan Menggunakan DBSCAN dan ST-DBSCAN.

Rahim, M. S. B. M., & Daman, D. (2006). The Development of Spatio-Temporal Data Model for Dynamic Visualization of Virtual Geographical Information System. Transportation, 1(January), 21–30. https://doi.org/10.1002/ejoc.201200111