JSTAR
Vol 5 No 2 (2025): Jurnal Statistika Terapan

KLASTERISASI USAHA PERTANIAN PERORANGAN TANAMAN PANGAN DI PROVINSI NUSA TENGGARA TIMUR: PERBANDINGAN ALGORITMA K-MEANS DAN K-MEDOIDS

Apriliani Gustiana (Badan Pusat Statistik)
Firrar Ayu Hastungkara Sudrajat (Badan Pusat Statistik)



Article Info

Publish Date
31 Dec 2025

Abstract

Agricultural Development is one of main goals outlined in the Dasa Cita of Nusa Tenggara Timur’s (NTT) Governor and Vice Governor. In line with that, advancing the agriculture, plantation, livestock, fisheries and maritime sectors as leading sectors that sustainable and based on regional potential is the first foundation stated in Program 7 Pilar NTT Government. This paper examines the clustering of Individual Agricultural Holdings (UTP) using Sensus Pertanian 2023 data on five predominant variety of food crops (dry land paddy, wet land paddy, maize, cassava, sweet potato) in order to reveal heterogeneity in food crop orientation by regencies/municipality in NTT and to inform targeted, evidence-based agricultural support. The methodology used for clustering are K-Means and K-Medoids which is then evaluated with Davies–Bouldin Index (DBI) and Silhouette Coefficient. The results showed that the optimal number of clusters in this study were four clusters. K-Medoids performs best (DBI = 0.86; silhouette = 0.40), slightly outperforming K-Means (DBI = 0.88; silhouette = 0.39). The resulting clusters can be differentiated into UTP Dryland Paddy, UTP Wetland Paddy, UTP Secondary Food Crops (maize, cassava, sweet potato), and Non-Concentration, offering actionable guidance for policy making.

Copyrights © 2025






Journal Info

Abbrev

JSTAR

Publisher

Subject

Humanities Computer Science & IT Economics, Econometrics & Finance Social Sciences

Description

Aim: JSTAR studies applied statistics at the regional and national levels of East Nusa Tenggara which are directed to contribute to the government in making regional development policies. JSTAR pays special attention to official and modeling statistics, big data and data mining, and the application ...