ANALISIS KLASTER KEMISKINAN KABUPATEN KOTA DI PROVINSI BANTEN MENGGUNAKAN METODE K-MEANS

Abstract

Poverty is a special problem that needs to be discussed in Indonesia. Because poverty that occurs in a certain area will have an impact on development in Indonesia. Banten province has a percentage of 6.04 percent with a total poverty of 576.62 thousand poor people. Given this, the Provincial Government of Banten requires knowledge of district/city socio-economic data in Banten Province. This study aims to cluster regencies/cities in Banten Province using the K-Means method. The K-Means method is a clustering technique that aims to group data based on the same characteristics. The variables observed in this study were the percentage of poor people aged 15 and over who had graduated from elementary/junior high school, the percentage of poor people aged 15 and over who had finished high school and above, the percentage of poor people aged 15 and over by district/city not working, the percentage of poor people aged 15 Years and over by district/city working in the informal sector, percentage of poor population aged 15 and over by district/city working in the formal sector, percentage of expenditure per capita for food by district/city and poor status. the results of the analysis showed that the district/city poverty clustering in Banten Province using the K-means method obtained 2 clusters, namely cluster 1 consisting of Tangerang City, Cilegon City and South Tangerang City and cluster 2 consisting of Pandeglang Regency, Lebak Regency, Tangerang, Serang Regency, Serang City. And formation 2 cluster has a good structural test value based on the results of the silhouette index so that 2 cluster has the best accuracy value compared to 3 clusters and 4 cluster.

Keywords

• silhouette index
• k-means
• kemiskinan
• klaster
• poverty
• cluster