Year: 2025
Type: article
Abstract: This study aims to classify the sales performance of culinary products at Tenant 1 of PT. Usaha Mitra Milenial using the K-Nearest Neighbors (KNN) algorithm. The research focuses on how normalization techniques influence the accuracy of the classification process. Data were collected from 150 recorded transactions containing attributes such as product name, transaction time, and sales amount. Two normalization methods, namely Min–Max Scaling and Z-Score Standardization, were applied to standardize the dataset before model training. The data were divided into training and testing sets using an 80:20 ratio, and model evaluation was conducted using accuracy and a classification report consisting of precision, recall, and F1-score. The experiment demonstrated that the KNN model with K = 5 achieved a classification accuracy of 100%, successfully distinguishing product sales into three categories: high, medium, and low. These findings show that data normalization significantly enhances model consistency and improves classification reliability. The application of this model provides valuable insight for business owners in optimizing inventory planning, pricing strategies, and sales forecasting.
Authors: Faiz Amrizal, Fatma Sari Hutagalung
Institution: Universitas Muhammadiyah Sumatera Utara
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4415543736