PENERAPAN ALGORITMA FUZZY C-MEANS UNTUK PENGELOMPOKAN HASIL PERTANIAN DI KECAMATAN SIPAHUTAR

  • Yusuf R.S. Silalahi Silalahi Program Studi Teknik Informatika, Fakultas Teknologi Industri, Institut Sains dan Teknologi TD Pardede, Medan, Indonesia
  • Swingly Purba Program Studi Teknik Informatika, Fakultas Teknologi Industri, Institut Sains dan Teknologi TD Pardede, Medan, Indonesia
  • Piala Mutiara Program Studi Teknik Informatika, Fakultas Teknologi Industri, Institut Sains dan Teknologi TD Pardede, Medan, Indonesia
Keywords: Fuzzy C-Means; clustering; data mining; agricultural products; Sipahutar District

Abstract

The agricultural sector is important in Sipahutar District, North Tapanuli Regency, yet
agricultural commodities exhibit substantial differences in land area, yield, and productivity. These
differences make objective, data-based grouping necessary for agricultural development planning.
This study applies the Fuzzy C-Means (FCM) algorithm to cluster agricultural products according
to their characteristics and to produce supporting information for agricultural development
decisions. The dataset consists of 108 annual observations from 18 commodities recorded during
2020–2025, using land area, harvested production, and productivity as numerical attributes. The
workflow includes data cleaning, Z-Score standardization, testing candidate cluster numbers,
iterative FCM computation, internal cluster validation, and interpretation. Testing two to seven
clusters selected three clusters. The selected model achieved an FPC of 0.8429, a Silhouette Score
of 0.5919, a Davies-Bouldin Index of 0.4176, an Xie-Beni Index of 0.1051, a Partition Entropy of
0.2673, and WCSS of 7.876. The main implementation log reached convergence at approximately
iteration 29. At the annual-observation level, six observations belonged to the high cluster, 37 to the
medium cluster, and 65 to the low cluster. At the dominant-commodity level, Pineapple was classified as high; Ginger, Turmeric, Lemongrass, Galangal, Cabbage, Water Spinach, and Chayote
were identified as commodities requiring monitoring or improvement; and most other horticultural
commodities were dominated by the low cluster. These findings demonstrate that FCM can map
agricultural heterogeneity and translate clustering results into data-based development
recommendations.

Published
2026-11-03
How to Cite
Silalahi, Y. R. S., Purba, S., & Mutiara, P. (2026). PENERAPAN ALGORITMA FUZZY C-MEANS UNTUK PENGELOMPOKAN HASIL PERTANIAN DI KECAMATAN SIPAHUTAR. IKRA-ITH Informatika : Jurnal Komputer Dan Informatika, 11(1), 9-14. Retrieved from https://journals.upi-yai.ac.id/index.php/ikraith-informatika/article/view/8026