Performance Evaluation of Machine Learning Algorithms for Supply Chain Data Classification

Maniah, Maniah (2026) Performance Evaluation of Machine Learning Algorithms for Supply Chain Data Classification. ESL (Engineering Scince Letter), 5 (1). pp. 34-40. ISSN 2961-872X

	Text 3. Bukti Korespondensi Jurnal ESL - edit 29102627.pdf Download (3MB)
	Text 2. Maniah_Similarity index versi terbit.pdf Download (2MB)

Abstract

Forecasting systems that are data-driven are of great importance in streamlining industrial and business processes during the digital transformation age. Supply chain management (SCM) is among the most significant processes for enhancing operational efficiency and supporting strategic decision-making. This study seeks to evaluate the performance of two machine learning-based classification algorithms, namely Naive Bayes and the k-Nearest Neighbours (K-NN) algorithm, using data in the supply chain. Some of the most valuable operational attributes, including payment method, customer segment, shipment status, profit per transaction, and customer location, are stored in the database. The data were first cleaned and then normalised and label-encoded, after which they were split into training and test sets with a ratio of 80:20. The performance of the two algorithms was assessed using accuracy, precision, recall, and F1-score. The findings of the research indicate that Naive Bayes is the most promising algorithm; its accuracy and precision are 99.75%, and its recall rate is close to 100% in the majority of the classes. These findings show that Naive Bayes is a probabilistic algorithm that better fits the data distribution than a distance-based K-NN algorithm

Item Type:	Article
Subjects:	Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	Article Dosen
Depositing User:	Maniah MN Maniah
Date Deposited:	03 Nov 2026 03:04
Last Modified:	03 Nov 2026 03:04
URI:	http://eprints.ulbi.ac.id/id/eprint/4329

Actions (login required)

View Item