Implementasi K-Means Clustering Kelompok Provinsi Penghasil Bawang Merah di Indonesia Tahun 2023

Abstract

Shallots are a strategic horticultural commodity that plays a vital role in food security and price stabilization in Indonesia. However, inter-provincial production data is still in the raw form, making it difficult to identify the distribution patterns of high, medium, and low-production areas. This study aims to implement the K-Means Clustering algorithm to group 38 provinces in Indonesia based on 2023 shallot production and productivity data sourced from the Central Statistics Agency (BPS). The method used is a data mining approach through the stages of data selection, preprocessing (Min-Max normalization), data transformation, determining the number of clusters using the Elbow method, and evaluation using the Silhouette Score with the assistance of Orange Data Mining software. The results show the formation of three clusters: a low-medium production cluster dominated by most provinces outside the main centers, a high production cluster consisting of Central Java and East Java as national centers, and a very high production cluster encompassing several provinces such as West Java and West Nusa Tenggara. The results of this grouping provide an overview of the inequality in production distribution between regions and can be the basis for formulating development policies, production equality, and shallot distribution planning in a more targeted and data-based manner