Regional Clustering of Indonesian Regencies and Municipalities Based on the Poverty Severity Index and the Prevalence of Inadequate Food Consumption in 2024 Using the K-Medoids
Keywords:
K-Medoids Clustering, Poverty Severity Index, Inadequate Food ConsumptionAbstract
Abstract. Indonesia's national development aims to improve public welfare, particularly by ensuring the fulfillment of basic food needs and reducing poverty. However, regional disparities, price fluctuations, and the impacts of climate change continue to create inequalities in food security across regions. This study aims to classify regencies and municipalities in Indonesia based on two key indicators, namely the poverty severity index and the prevalence of inadequate food consumption in 2024, using the K-Medoids clustering algorithm. This method was selected because of its robustness in handling data containing outliers compared with mean-based clustering methods. Data obtained from Statistics Indonesia (BPS) were preprocessed through missing value assessment, outlier detection, and data standardization using the Standard Scaler to ensure reliable analytical results. The findings indicate that the optimal clustering solution consists of two clusters, as determined by the highest Silhouette Index value of 0.67. The iterative clustering process identified Tambrauw Regency as the medoid representing Cluster 1 and Tana Toraja Regency as the medoid representing Cluster 2. The resulting cluster distribution is highly imbalanced, with Cluster 1 comprising only 87 regencies/municipalities, while the remaining 427 regencies/municipalities are grouped into Cluster 2. In terms of characteristics, Cluster 1 represents regions with high levels of both the poverty severity index and the prevalence of inadequate food consumption, whereas Cluster 2 consists of regions with relatively more stable and favorable socioeconomic conditions. These findings provide valuable evidence for supporting data-driven regional development policies aimed at promoting more equitable welfare across Indonesia.
References
Badan Pusat Statistik, Indikator Kesejahteraan Rakyat 2024, vol. 53. Badan Pusat Statistik, 2024.
M. Taufiqurrahman, “Analisis Indeks Keparahan Kemiskinan di Pulau Jawa 2012-2021,” Jurnal Ilmu Ekonomi (JIE), vol. 6, no. 4, pp. 621–634, 2022, doi: https://doi.org/10.22219/jie.v6i4.23011.
K. Rahakbauw and P. Lindiasari Samputra, “Analisis Tantangan dan Strategi Ketahanan Pangan di Indonesia,” Jurnal Agrica, vol. 18, no. 1, 2025, doi: 10.31289/agrica.v18i1.11883.
Badan Pangan Nasional, Indeks Ketahanan Pangan. Badan Pangan Nasional, 2023.
H. Hafid, S. M. Meliyana, I. Muthahharah, and Z. Mar’ah, “Implementation K-Medoids Algorithm for Clustering Indonesian Provinces by Poverty and Economic Indicators,” Quantitative Economics and Management Studies, vol. 6, no. 2, pp. 219–225, Apr. 2025, doi: 10.35877/454ri.qems3940.
J. Suryadi, “Clustering Ketidakcukupan Konsumsi Pangan Per Kabupaten/Kota Menggunakan Algoritma K-Means dan Fuzzy C-Means,” Jurnal Komputer dan Informatika, vol. 19, no. 1, pp. 30–35, 2024.
Yosia and B. Siregar, “Comparative Analysis of K-Means and K-Medoids Algorithms for Product Sales Clustering and Customer,” Journal of Mathematics, Computations and Statistics, vol. 7, no. 2, pp. 360–370, Oct. 2024, doi: 10.35580/jmathcos.v7i2.4053.
Badan Pusat Statistik, “Indeks Keparahan Kemiskinan (P2) Menurut Kabupaten/Kota, 2024,” Badan Pusat Statistik. Accessed: Jul. 31, 2026. [Online]. Available: https://www.bps.go.id/id/statistics-table/2/NjIzIzI=/indeks-keparahan-kemiskinan-p2-menurut-kabupaten-kota.html
Badan Pusat Statistik, “Prevalensi Ketidakcukupan Konsumsi Pangan (Persen) Per Kabupaten/kota (Persen), 2024,” Badan Pusat Statistik. Accessed: Jul. 31, 2026. [Online]. Available: https://www.bps.go.id/id/statistics-table/2/MjI2OSMy/prevalensi-ketidakcukupan-konsumsi-pangan-persen-per-kabupaten-kota.html
S. Soesmono, R. Pertiwi, B. Saputri, N. Putri, and E. Widodo, “Pengelompokan Provinsi di Indonesia Berdasarkan Tingkat Pengangguran Tahun 2023 Menggunakan K-Medoids,” Emerging Statistics and Data Science Journal, vol. 3, no. 1, 2025.
A. ur Rehman and S. B. Belhaouari, “Unsupervised Outlier Detection in Multidimensional Data,” J. Big Data, vol. 8, no. 1, Dec. 2021, doi: 10.1186/s40537-021-00469-z.
M. Sriningsih, D. Hatidja, and J. D. Prang, “Penanganan Multikolinearitas dengan Menggunakan Analisis Regresi Komponen Utama pada Kasus Impor Beras Di Provinsi Sulut,” Jurnal Ilmiah Sains, vol. 18, 2018, doi: https://doi.org/10.35799/jis.18.1.2018.19396.
J. Han, M. Kamber, and J. Pei, Data Mining. Concepts and Techniques, 3rd Edition (The Morgan Kaufmann Series in Data Management Systems), 3rd ed. Morgan Kaufmann Publishers, 2011.
U. Kumar and N. Sharma, “An Improved Hybrid_Stacked Deep Neural Network (HDNN) Model for Enhanced Weather Forecasting,” Journal of Information Systems Engineering and Management, vol. 10, no. 15s, pp. 2468–4376, 2025, doi: https://doi.org/10.52783/jisem.v10i15s.2468.
V. Fauziah and Suliadi, “Pengelompokkan Desa di Jawa Barat Berdasarkan Variabel Penyusun Indeks Desa Membangun,” Bandung Conference Series: Statistics, vol. 4, no. 1, pp. 221–229, Feb. 2024, doi: 10.29313/bcss.v4i1.11589.
G. Vardakas, I. Papakostas, and A. Likas, “Deep Clustering Using the Soft Silhouette Score: Towards Compact and Well-Separated Clusters,” Feb. 2024, [Online]. Available: http://arxiv.org/abs/2402.00608
A. P. Ayuni, D. Kusnandar, and S. Martha, “Implementasi Algoritma K-Medoids dan Clustering Large Applications (Clara) Dengan Optimasi Silhouette Coefficient (Studi Kasus: Pengelompokan Indeks Pembangunan Manusia Berdasarkan Kabupaten/Kota di Indonesia),” Buletin Ilmiah Math. Stat. dan Terapannya (Bimaster), vol. 13, no. 2, pp. 191–200, 2024, doi: https://doi.org/10.26418/bbimst.v13i2.76959.
