METODE NAÏVE BAYES UNTUK KLASIFIKASI KESETIAAN PELANGGAN TERHADAP RESTORAN JOKO SOLO PADA SISTEM CRM (CUSTOMER RELATIONSHIP MANAGEMENT)

Abstract

Increasing competition in the culinary business sector requires companies not only to acquire new customers but also to maintain the loyalty of existing customers. One approach to achieving this objective is the utilization of customer data through a Customer Relationship Management (CRM) system integrated with data mining techniques. This study aims to implement the Naïve Bayes method for customer loyalty classification at Joko Solo Restaurant and to analyze its role in supporting data-driven decision making. The study employed a quantitative approach using a data classification method. The dataset consisted of customer visit frequency, transaction value, and complaint history obtained from the restaurant’s CRM system. These data were processed using the Naïve Bayes algorithm to determine customer loyalty categories based on the highest probability value generated by the model. The results indicate that the system successfully classified customers into three loyalty categories: loyal customers (60%), moderately loyal customers (36%), and non-loyal customers (4%). The findings demonstrate that the Naïve Bayes method is capable of transforming historical customer data into valuable information that enables restaurant management to identify customer loyalty levels more objectively. The integration of Naïve Bayes into the CRM system supports the development of marketing strategies, service quality improvement, and customer retention efforts more effectively. This study contributes to the advancement of data mining implementation in CRM systems, particularly within the culinary industry, by providing a customer loyalty classification model that can be utilized as a foundation for data-driven managerial decision making.