K-Means Algorithm with Davies Bouldin Criteria for Clustering Provinces in Indonesia Based on Number of Events and Impacts of Natural Disasters

  • Yuli Asriningtias UTY
  • Joko Aryanto Universitas Teknologi Yogyakarta
Keywords: Natural disasters, K-Means Clustering, Davies Bouldin Index, Dataset

Abstract

Indonesia has 34 Provinces with a geographical position at the meeting zone of the Eurasian, Indo-Australian, Pacific and Philippine slab, this makes Indonesia vulnerable to the threat of geological disasters. Other disaster threats arise due to climate change and people's bad behavior towards the environment which has an impact on natural and environmental damage. Based on data natural disasters and their impacts over the last five years, a study was conducted to find province clusters that fall into disaster-prone criteria, the number of disaster victims and the impact on building damage caused. The research was conducted using rapidminer tools with the K-Means Clustering algorithm and optimization using the Davies Bouldin Index (DBI). The method used is collecting datasets, preprocessing data, modeling and analyzing DBI. The results, found that the cluster of disaster-prone provinces was East Java, followed by Central Java and West Java. The large number of disaster victims are the provinces of Lampung and West Nusa Tenggara. The biggest impact of damage buildings is the Province of West Nusa Tenggara

 

Submitted
2022-05-23
Accepted
2022-07-30
How to Cite
Asriningtias, Y., & Aryanto, J. (2022). K-Means Algorithm with Davies Bouldin Criteria for Clustering Provinces in Indonesia Based on Number of Events and Impacts of Natural Disasters. International Journal of Engineering Technology and Natural Sciences, 4(1). https://doi.org/10.46923/ijets.v4i1.147