Self Organizing Maps Method for Grouping Provinces in Indonesia Based on the Landslide Impact

Authors

  • Suwanda Risky Universitas Negeri Padang
  • Syafriandi
  • Dony Permana
  • Dina Fitria

DOI:

https://doi.org/10.24036/ujsds/vol1-iss3/15

Keywords:

Self Organizing Maps, Landslide, Cluster Validation

Abstract

Indonesia is a disaster-prone country due to its climatic, soil, hydrological, geological, and geomorphological conditions. A disaster is an event or chain of events that threatens and disrupts people's lives and livelihoods. A natural disaster is a disaster caused by an event or series of events caused by nature such as a landslide. The number of landslide disaster events in Indonesia varies from province to province, this is due to differences in the characteristics of each province in Indonesia. So that the impact caused by the landslide disaster is also different. Therefore, it is necessary to group and profile so that it can be known which province has the largest impact on landslide disasters. This study used the Self Organizing Maps method in a grouping. The number of clusters to be formed is 3 based on the optimal value of internal cluster validation (Dunn, Connectivity, and Silhouette Index). Cluster 1 consists of 31 provinces, and the average impact of landslides is small. In cluster 2 consisting of 2 provinces, there are 4 dominantly more significant impacts. Cluster 3 consisting of 1 province has 1 dominant impact greater. So it can be concluded that most provinces in Indonesia have a relatively small impact on landslide disasters. However, some provinces have a very large impact on landslides, namely the provinces of West Java, Central Java, and East Java.

Published

2023-05-31

How to Cite

Suwanda Risky, Syafriandi, Dony Permana, & Dina Fitria. (2023). Self Organizing Maps Method for Grouping Provinces in Indonesia Based on the Landslide Impact . UNP Journal of Statistics and Data Science, 1(3), 180–187. https://doi.org/10.24036/ujsds/vol1-iss3/15

Most read articles by the same author(s)

1 2 3 4 5 6 > >>