Klasifikasi Loyalitas Pengguna Shopee Pada Mahasiswa Generasi Z Dengan Algoritma Learning Vector Quantization
Year: 2026
Type: article
Abstract: Generation Z shoppers on Shopee switch platforms easily whenever another marketplace offers better promotions or faster service. This study classifies the loyalty level of Shopee customers among Generation Z students at the Faculty of Engineering, Universitas Asahan, using Learning Vector Quantization (LVQ). An online questionnaire was answered by 105 students and 101 of them were active Shopee users. Six input variables were measured on a 1 to 5 Likert scale: purchase frequency, customer satisfaction, transaction security, service quality, promotional influence and repurchase intention. Loyalty was grouped into low, medium and high classes. The data were normalized with Min-Max and split 80:20 by stratified sampling into 81 training and 20 testing records. Training used a learning rate of 0.05, a decay constant of 0.1 and 10 epochs. The model correctly classified 16 of the 20 test records for an accuracy of 80%. Every high and low loyalty record was recognized, while four of the seven medium records were predicted as high. The final weights show that repurchase intention separates the classes far more clearly than purchase frequency. The model was deployed in a PHP and MySQL web application for data management, training, testing and classification of new respondents.
DOI: 10.54314/vzwzj963
Authors: Laura Gusti Ayunda, Bambang Irwansyah
Institution: Universitas Asahan
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7228535279
