Analysis of Purchasing Trends and Product Recommendations of Customer Clusters for Each Product Using Bayesian Network

In micromarketing, data mining is actively conducted using large-scale data to understand customers' detailed needs. Customer Relationship Management aims to improve customer satisfaction and increase revenue using analytical techniques such as RFM analysis. In this study, to segment customers and improve the accuracy of RFM analysis, we use a Bayesian network to analyze the causal relationships between purchasing trends for each customer cluster and product category. For products in categories where causal relationships have been confirmed, a Bayesian network is used to clarify detailed causal relationships between products. Based on sales history data, the top 20% customers with regard to purchase frequency and purchase amount were classified using cluster analysis, and the causal relationships were analyzed. The results revealed differences in the products sold together in each cluster, suggesting the potential for detailed analysis of purchasing trends through customer segmentation.