Claim Missing Document
Check
Articles

Found 1 Documents
Search

Perbandingan Metode Agglomerative Hierarchical Clustering dengan K- Medoids Clustering dalam Pengelompokan Tipologi Pembangunan Daerah Kabupaten/Kota Provinsi Jawa Barat Arina Maizar Hafid; Nisfatun Nurul Ana; Sabila Hardiana; Rahayu Setyaningsih; Siti Claudia Tane; Yazida Salsabila; Fatkhurohman Fauzi
Seminar Nasional Official Statistics Vol 2026 No 1 (2026): Seminar Nasional Official Statistics 2026
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34123/semnasoffstat.v2026i1.2829

Abstract

Sustainable regional development requires appropriate mapping of regional typologies to help minimize socio-economic disparities. West Java Province was selected because its 27 regencies/cities exhibit diverse development characteristics and heterogeneous socio-economic conditions, necessitating regional clustering to identify patterns of development typology. This study aims to identify the development typology of regencies/cities in West Java Province in 2025 and compare the performance of two clustering methods, namely *Agglomerative Hierarchical Clustering* (AHC) and *K-Medoids Clustering*. The secondary data were obtained from the Central Statistics Agency (BPS) of West Java Province for 2025, comprising 27 regencies/cities and six development indicator variables. The data were standardized using the Z-score method and assessed for suitability through multicollinearity examination and the Kaiser-Meyer-Olkin (KMO) test (KMO = 0.8209) prior to clustering. The results showed that AHC with average linkage produced two optimal clusters consisting of 18 and 9 regencies/cities, with an agglomerative coefficient of 0.7659, while K-Medoids produced two clusters consisting of 17 and 10 regencies/cities. The Silhouette Coefficient evaluation indicated that AHC performed better, with a score of 0.404 compared to 0.3773 for K-Medoids. Therefore, AHC was selected as the best-performing clustering method. The resulting clustering is expected to serve as a reference for local governments in formulating more targeted development policies.