Analisis Clustering K-Means untuk Pemetaan Tingkat Pengangguran Terbuka di Provinsi-Provinsi Indonesia Tahun 2013-2023

Abstract

This study analyzes unemployment rates in Indonesian provinces using data from the Central Statistics Agency (BPS) for the period 2013-2023 and the K-Means clustering algorithm. The aim is to group regions based on the Open Unemployment Rate (TPT). Two main clusters were produced: one with a high unemployment rate (cluster 0) and one with a low unemployment rate (cluster 1). Cluster 0 consists of 12 provinces, while cluster 1 consists of 22 provinces. The model evaluation shows a Davies-Bouldin Index score of 0.7041, indicating good clustering quality. The clustering results are visualized in the form of a map for easy interpretation. This research is expected to help policymakers design more effective policies in reducing unemployment in Indonesia, provide deep insights into regional differences in terms of unemployment, and support targeted decision-making.