IMPLEMENTASI K-MEANS CLUSTERING UNTUK MENINJAU KEPUASAN PELANGGAN TERHADAP LOYALITAS PELANGGAN DI OPTIK AKSARA MENGGUNAKAN C4.5

Abstract

Optik Aksara faces difficulties in understanding customer satisfaction and loyalty because questionnaire data has not been optimally analyzed. This study aims to cluster customer satisfaction levels and predict customer loyalty using a data mining approach. The methods used are K-Means Clustering for grouping customer satisfaction and Decision Tree C4.5 for customer loyalty classification. Data were collected through questionnaires based on five service quality dimensions: tangible, reliability, responsiveness, assurance, and empathy. The research stages include data collection, preprocessing, clustering, classification, and desktop-based system implementation. The results show that K-Means successfully groups customers into satisfaction categories, while C4.5 generates decision rules to predict customer loyalty. The developed system assists management in data-driven decision making to improve service quality and customer retention.

Key Words: Data Mining, K-Means, C4.5, Customer Satisfaction, Loyalty