Implementasi Data Mining Untuk Pengelompokan Pengangguran Terbuka di Indonesia Dengan Metode Clustering

Abstract

The number of unemployed as of August 2021 was 9.1 million people from all types of educational circles in Indonesia, so the percentage of unemployed in Indonesia is quite large and is a factor that can be underlined. The unemployment rate will increase every year if there are no significant efforts to reduce the percentage of unemployed in Indonesia. Therefore, there is a need for government realization in providing jobs in accordance with the groupings in the territory of Indonesia in order to know the level of unemployment clusters of the Indonesian population. The implementation of data mining using the clustering method in this study aims to determine groups of the unemployed based on the level of unemployment in the territory of Indonesia. The clustering method using the k-means algorithm can group Indonesia into three clusters, namely cluster 1 is a group of regions with a low unemployment rate which includes 3 regions, cluster 2 is a group of regions with a moderate unemployment rate which includes 10 regions, and cluster 3 is a group regions with high unemployment rates covering 21 regions. The results of the analysis obtained the percentage of each cluster, namely cluster 1 has 9% of the area, cluster 2 has 29% of the area, and cluster 3 has 62% of the area. These results are expected to provide information to the community and local government, especially in areas with high unemployment rates, to be able to carry out ideas to reduce the percentage rate of unemployment in Indonesia.

Keywords: Data mining, Clustering, K-means Algorithm