IMPLEMENTASI METODE K-MEANS DAN K-MEDOIDS PADA PENGELOMPOKAN PROVINSI INDONESIA BERDASARKAN ASPEK PENDIDIKAN PEMUDA

Tusyakdiah, Halima
Hasanah, Insani
Panggol, Sri Arista
Ramdhanti, Tiara
Permatasari, Retno
Cusanti, Cusanti
Widodo, Edy

Community Services and Social Work Bulletin
Vol 3, No 1 (2023): Community Services and Social Work Bulletin Volume 3 No. 1 2023

The quality of education in Indonesia is still a concern, seen from a number of problems that become obstacles to improving the quality of education as well as affecting the quality of Indonesian youth. This study aims to group provinces in Indonesia based on the aspect of youth education using the K-Means and K-Medoids methods. To determine the optimum k, the average silhouette method is used and the SW and SB ratio is used to evaluate the cluster results. The results obtained are 2 clusters optimum. For the K-Means method, cluster 1 consists of 19 provinces and cluster 2 consists of 14 provinces. Whereas in the K-Medoids method, cluster 1 consists of 22 provinces and cluster 2 consists of 11 provinces. The K-Means method is better than the K-Medoids method because it has a ratio value of 0.527941 which is smaller than the K-Medoid ratio value of 0.5612719.Keyword: K-Means; K-Medoid; Education; Average Silhouette; Standard Deviation.