Analysis of Unemployment Patterns in Indonesia Using K-Means Clustering and Identification of Dominant Factors Using Random Forest
Abstract
The disparity in the unemployment rate between provinces in Indonesia, exacerbated by the COVID-19 pandemic, is the main focus of this study. This study aims to (1) map the grouping of unemployment patterns in 34 provinces based on the Open Unemployment Rate (TPT) time series data for the 2020-2024 period, and (2) analyze the most significant socio-economic determinants of the formation of these patterns. Applying a two-stage methodology, cluster analysis using K-Means—validated through the Elbow Method and Silhouette Score of 0.456—succeeded in classifying the provinces into four different groups. The two prominent clusters identified were "Cluster 2: Pandemic Shock Pattern" (e.g., DKI Jakarta, West Java) which showed a surge in TPT above 10%, and "Cluster 3: Resilient Pattern" (e.g., Bali, DIY) which showed the lowest TPT rate and fastest recovery. Furthermore, the Random Forest Classifier analysis identified a hierarchy of determining factors, with the 2024 Average School Length (RLS) as the strongest predictor, followed by the 2024 Provincial Minimum Wage (UMP) and the 2024 GDP. These findings underline that the quality of human capital (education) is a more crucial factor than economic output (GDP) in shaping the resilience of the labor market. The study concludes the need for differentiated and cluster-specific unemployment policy interventions, rejecting a nationally uniform approach.
Downloads
References
K. Y. Mahendra, M. Susilawati, N. Luh, and P. Suciptawati, “MEMODELKAN TINGKAT PENGANGGURAN TERBUKA,” vol. 10, no. 1, pp. 20–25, 2021.
R. Maliki, K. Falgenti, S. Priani, F. Fithri, M. Suherman, and D. S. Nugraha, “Perbandingan Tingkat Pengangguran Terbuka Provinsi di Indonesia Berbasis Metode K-Means Clustering,” vol. 2, no. 2, pp. 109–116, 2022.
H. D. Dermawan, R. Kurniawan, and Y. A. Wijaya, “CESS Analisis Data Penjualan Menggunakan Algoritma K-Means Clustering Pada Toko Daun Indah Di Shopee Analysis of Sales Data Using K-Means Clustering Algorithm at Daun Indah Shop in Shopee,” vol. 9, no. 1, pp. 175–191, 2024.
E. U. Oti, M. O. Olusola, F. C. Eze, and S. U. Enogwe, “Comprehensive Review of K-Means Clustering Algorithms,” vol. 7, no. 8, pp. 22–23, 2021, doi: 10.31695/IJASRE.2021.34050.
G. H. Fatimah and G. E. Setyowisnu, “Pengelompokan Kabupaten / Kota di Provinsi Jawa Barat Berdasarkan Tingkat Pengangguran Terbuka Tahun 2023 Menggunakan K- Means Clustering,” vol. 11, no. 2, pp. 279–291, 2025.
J. E. Simarmata, D. Chrisinta, and M. Purnomo, “Journal of Research in Mathematics Trends and Technology Implementation of K-Means Clustering to Human Development Indicators in East Nusa Tenggara,” vol. 6, no. 2, pp. 46–56, 2024, doi: 10.32734/jormtt.v6i2.17066.
A. I. Ramadhan, P. D. Atika, and K. F. Ramdhania, “Analisis Clustering K-Means untuk Pemetaan Tingkat Pengangguran Terbuka di Provinsi-Provinsi Indonesia Tahun 2013-2023,” vol. 5, no. 2, pp. 109–122, 2025.
D. Nishfi, I. Huda, C. Prianto, and R. M. Awangga, “Analisis Sentimen Perbandingan Layanan Jasa Pengiriman Kurir Pada Ulasan Play Store Menggunakan Metode Random Forest dan Descision Tree,” no. 54, 2023.
K. Angkatan et al., “Workforce Classification in West Java 2018 With Random Forest,” vol. 17, no. 2, pp. 240–251, 2021, doi: 10.20956/jmsk.v17i2.11680.
P. S. Matematika, F. Matematika, D. A. N. Ilmu, P. Alam, and U. N. Jakarta, “PEMODELAN GEOGRAPHICALLY WEIGHTED REGRESSION MENGGUNAKAN PEMBOBOT KERNEL FIXED DAN ADAPTIVE PADA KASUS,” 2021.
Copyright (c) 2026 Yuliana , Rima Fazri Ramadhania, Asep Abdul Latip, Muhammad Farhan Harahapa
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
