<![CDATA[The rapid growth of retail transaction data has created new opportunities for businesses to analyze customer purchasing behavior and improve decision-making strategies. Market basket data contains valuable information about product combinations purchased together within a single transaction, which can reveal hidden patterns of consumer behavior. This study aims to apply the K-Means clustering algorithm to segment market basket transaction data based on similarities in purchasing patterns. The research method involves several stages, including data preprocessing, transformation of transaction data into a binary feature matrix, determination of the optimal number of clusters, and clustering analysis using the K-Means algorithm. The results show that the clustering process successfully groups transactions into several clusters representing different purchasing characteristics. Each cluster reflects distinct consumer behavior patterns such as routine household purchases, breakfast-related items, snack-oriented transactions, and fresh product selections. These findings demonstrate that K-Means clustering can effectively identify meaningful patterns within market basket datasets. The clustering results provide useful insights that can support retail strategies such as targeted promotions, product bundling, store layout optimization, and inventory management. Overall, the application of clustering techniques in market basket analysis contributes to improving data-driven decision-making and enhancing the understanding of customer purchasing behavior in retail environments.]]>
Copyrights © 2026
