semanTIK
Vol. 11 No. 2 (2025): SemanTIK : Teknik Informasi

Sistem Pemilihan Rekomendasi Produk UMKM Kopi menggunakan Metode K-Nearest Neigbors

M. Ardiansyah (Politeknik Negeri Sriwijaya)
Ahmad Taqwa (Politeknik Negeri Sriwijaya)
Ade Silvia Handayani (Politeknik Negeri Sriwijaya)



Article Info

Publish Date
16 Dec 2025

Abstract

Seiring dengan meningkatnya tren konsumsi kopi di berbagai kalangan dan semakin beragamnya preferensi konsumen terhadap produk kopi, hal ini juga dipicu oleh banyaknya produk kopi bermunculan di pasaran dengan berbagai varian harga, jenis olahan, dan asal kopi yang ditawarkan oleh pelaku Usaha Mikro, Kecil, dan Menengah (UMKM). Semakin banyaknya varian produk yang beredar membuat konsumen semakin sulit menemukan produk kopi yang sesuai dengan preferensi mereka. Metode K-Nearest Neighbors (KNN) menawarkan pendekatan yang efektif untuk membantu konsumen dalam menemukan rekomendasi produk kopi. Penelitian ini mengimplementasikan metode KNN mengukur jarak kedekatan antara preferensi pengguna dan karakteristik produk menggunakan dua metrik pengukuran, yaitu Euclidean dan Manhattan. Hasil evaluasi pengujian menunjukkan bahwa metrik jarak Euclidean memberikan tingkat akurasi tertinggi sebesar 92.2%, diikuti oleh Manhattan sebesar 91.8%. Berdasarkan hasil tersebut, Euclidean merupakan pilihan optimal dalam sistem rekomendasi yang dikembangkan, terbukti mampu memberikan rekomendasi produk kopi kepada konsumen dengan tingkat akurasi mencapai 92,2%. Along with the increasing trend of coffee consumption across various demographics and the growing diversity of consumer preferences for coffee products, this is also driven by the multitude of coffee products emerging in the market with various price ranges, types of processing, and origins offered by micro, small, and medium enterprises (MSMEs). The increasing variety of products available makes it more difficult for consumers to find coffee products that match their preferences. The K-Nearest Neighbors (KNN) method offers an effective approach to help consumers find coffee product recommendations. This research implements the KNN method to measure the proximity between user preferences and product characteristics using two measurement metrics, namely Euclidean and Manhattan. The evaluation results show that the Euclidean distance metric provides the highest accuracy level of 92.2%, followed by Manhattan at 91.8%. Based on these results, Euclidean is the optimal choice in the developed recommendation system and has proven capable of providing coffee product recommendations to consumers with the best recommendation results.

Copyrights © 2025






Journal Info

Abbrev

journal

Publisher

Subject

Computer Science & IT Control & Systems Engineering

Description

Jurnal semanTIK is a is one of the media publication of research results in the field of information technology. semanTIK is published Biannually, January-June and July-December and provide scientific publication medium for researchers, engineers, practitioners, academicians, and observers in the ...