Open AccessDetermine the clustering of cities in Indonesia for disaster management using K-Means by excel and RapidMiner
Author(s) -
Rienna Oktarina,
Junita Junita
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/794/1/012094
Subject(s) - natural disaster , index (typography) , cluster analysis , geography , indonesian , emergency management , business , risk management , agency (philosophy) , population , environmental health , economic growth , computer science , meteorology , medicine , sociology , economics , finance , linguistics , philosophy , world wide web , social science , machine learning
The impact of disasters can disrupt people’s lives, both natural and non-natural, resulting in human casualties, environmental damage, property loss, and psychological impact. Besides that, disasters that occur can also cause damage to health facilities, worship, education, and damage to homes, both severely, moderately, and lightly. The impact of disasters is so large, so a logistics warehouse is needed to handle the disaster. One of the countries prone to disasters, Indonesia which has the fourth largest population in the world with 34 provinces and 502 regions or cities. The purpose of this research is to determine the clustering of areas in Indonesia with a very high-risk, high-risk, moderate risk, low risk, and very low risk of disaster based on disaster data in Indonesian National Agency for Disaster Managementin 2010-2019 using K-Means calculations by Excel and the RapidMiner application. The results of both clustering methods are 6 cities that have a very high-risk index, 79 cities that have a high-risk index, 29 cities that have a medium risk index, 19 cities that have a low-risk index, and 369 cities have a very low-risk index. Thisresult can be considered for the construction of logistics warehouses for disaster management and K-Means method also can be used to know the clustering risk.