Unveiling Consumer Behavior Patterns Using K-Medoids and Association Rule: A Study on a Medium-Sized Grocery Store

This research addresses the problem of understanding consumer behavior and optimizing product bundling strategies in the context of a medium-sized grocery store. The objective is to explore the potential of association rule mining and K-Medoids clustering techniques for gaining insights into consumer preferences and identifying effective product bundling opportunities. To achieve this goal, we analyze real transaction data comprising 50 products and over 6,000 consumers. We apply association rule analysis to uncover frequent itemsets and association rules, which reveal patterns of correlated purchases among different products. This analysis enables us to identify significant associations, such as the relationship between eggs and instant noodles, and provides valuable insights into consumer purchase behavior. In addition, we employ K-Medoids clustering to group consumers based on their purchasing characteristics. By clustering consumers with similar preferences, we can identify distinct customer segments and tailor marketing strategies accordingly. This clustering approach enhances our understanding of customer behavior by grouping individuals who exhibit similar purchasing patterns and preferences. The integration of association rule mining and clustering techniques allows us to comprehensively analyze the data and gain deeper insights into consumer behavior. We uncover significant associations among products and identify customer segments with similar preferences. These findings provide actionable insights for retail businesses seeking to optimize product bundling strategies and enhance customer satisfaction.