Implementation K-Medoids Algorithm for Clustering Indonesian Provinces by Poverty and Economic Indicators

Hardianti Hafid; Sitti Masyitah  Meliyana; Isma  Muthahharah; Zakiyah  Mar’ah

doi:10.35877/454RI.qems3940

Hardianti Hafid Universitas Negeri Makassar (ID)
Sitti Masyitah Meliyana Universitas Negeri Makassar (ID)
Isma Muthahharah Universitas Negeri Makassar (ID)
Zakiyah Mar’ah Universitas Negeri Makassar (ID)

Keywords: K-Medoids, Poverty, Gross Regional Domestic Product (GRDP), Human Development Index (HDI), Regional Development

Viewed = 0 time(s)

Abstract

Regional development disparities in Indonesia remain one of the main challenges in formulating national development policies. This study aims to classify the 38 provinces in Indonesia based on four key indicators: the percentage of the population living in poverty, Gross Regional Domestic Product (GRDP) per capita, the open unemployment rate, and the Human Development Index (HDI), using the K-Medoids algorithm. This method was chosen due to its robustness to outliers and its ability to produce representative clusters. The data used are secondary data obtained from the Central Bureau of Statistics (BPS). The analysis process began with data standardization, determination of the optimal number of clusters using the Elbow and Silhouette methods, followed by clustering implementation and result interpretation. The analysis results identified four main clusters with distinct socioeconomic characteristics. Cluster 1 reflects provinces with moderate conditions, Cluster 2 represents more developed provinces, Cluster 3 highlights regions facing significant development challenges, and Cluster 4 consists of provinces with the most underdeveloped socioeconomic conditions. These findings indicate that the K-Medoids algorithm is effective in identifying inter-provincial disparity patterns and can serve as a foundation for formulating more targeted and inclusive development policies.

References

Adinata, A., Siahaan, T. E., & Wulandari, S. P. (2024). Evaluasi Faktor Sosial dan Kemiskinan di Provinsi Papua : Pendekatan Analisis Komponen Utama. 2(November), 195–203.

Alisjahbana, A. S., & Murningningtyas, E. (2018). Tujuan Pembangunan Berkelanjutan di Indonesia : Konsep Target dan Strategi Implementasi. Unpad Press.

Aqsa, F., & Masbar, R. (2023). Kausalitas Pertumbuhan Ekonomi Kepadatan Penduduk dan Kemiskinan di Provinsi Aceh. Jurnal Ilmiah Mahasiswa Ekonomi Pembangunan (JIM EKP) Fakultas, 8(1), 33–46.

Azmi, M., Putra, A. A., Vionanda, D., & Salma, A. (2023). Comparison the Performance of K-Means and K-Medoids Algorithms in Grouping Regencies / Cities in Sumatera Based on Poverty Indicators. 1, 59–66. https://doi.org/10.24036/ujsds/vol1-iss2/25

BRIDA. (2023). Rekonstruksi Strategi Penanggulangan Kemiskinan Jawa Tengah : Bukan Sekedar Bantuan Sosial. E-Journal Jateng, 1(1).

Budi, R. F. S., Widodo, W. L., & Salimy, D. H. (2014). Pengelolaan sumber daya energi di kalimantan untuk mendukung kemandirian energi dan pertumbuhan industri. Jurnal Pengembangan Energi Nuklir, 16(1), 43–53.

Dwiana, F. I., & Bahri, S. (2024). Implementasi Algoritma K-Medoids pada Klasterisasi Produksi Jenis Tanaman Pangan di Provinsi Jawa Timur. 20(1), 25–34. https://doi.org/10.24198/jmi.v20.n1.47278.25-34

Hafid, H., & Annisa, S. (2025). Implementation of K-Medoids and K-Prototypes Clustering For Early Detection of Hypertension. 19(1), 465–476.

Haumahu, G., & Matdoan, M. Y. (2022). Algoritma K-Medoids Clustering untuk Mengelompokkan Tingkat Kemiskinan Pada Kabupaten dan Kota di Kepulauan Maluku dan Papua ( K-Medoids Clustering Algorithm for Classification of Poverty Levels in Districts and Cities in The Maluku Islands and Papua ). Variance: Jourbal of Statsitics and Its Applications, 4(2), 81–87.

Ibrahim, R. N., Hayati, M. N., & Amijaya, F. D. T. (2020). Penerapan Algoritma K-Medoids pada Pengelompokan Wilayah Desa atau Kelurahan di Kabupaten Kutai Kartanegara (Studi Kasus : Data Hasil Pendataan Potensi Desa (PODES) Tahun 2018). Jurnal EKSPONENSIAL, 11(2), 153–158.