CLUSTERING KABUPATEN BERDASARKAN LUAS HUTAN MENGGUNAKAN METODE K-MEANS DI PROVINSI JAWA TENGAH

Abstract

Indonesia is one of the countries with the largest forest in the world. The tropical climate and high rainfall cause a lot of biodiversity in Indonesia’s forests. The existence of these forests can be utilized by many parties, both the government and the community in accordance with their functions to improve welfare. The government through the Central Statistic Agency has provided data information related to the forest area in various regions, one of which is Central Jawa Province but still requires development to obtain important information in the data. This study aims to divide the district based on forest area including protected forest, protected area, area for production, and area for other users in Central Java province using the K-Means Data Mining method. The data is obtained from Central Statistic Agency for the Central Java area, where four types of forest are to be grouped. The results of this study indicate that the grouping of districts based on the area of forest owned is based on the smallest Davies Bouldin (DB), which is 0.436 in the grouping with 2 clusters. The two clusters are distinguished based on the value of the proximity of the forest type attribute with the centroid point in each cluster. The clustering process grouped 26 districts in the province of Central Java into cluster 1, while cluster 2 consisted of 3 districts in Central Java, namely Grobogan, Blora, and Brebes districts.

Keywords: clustering, forests, K-Means