<![CDATA[With the rapid advancement of digital technology, companies are faced with an increasing volume and complexity of customer data. This necessitates a more effective approach to customer segmentation in order to better understand consumer behavior patterns and preferences. The large number of customers conducting transactions daily makes it challenging for managers to distinguish between highly frequent shoppers and less frequent ones based on transaction data and purchasing preferences. Therefore, the role of the K-Means clustering algorithm is crucial in addressing this issue. The objective of this study is to implement the K-Means Clustering algorithm for customer segmentation by grouping customers into three clusters: highly frequent shoppers, moderately frequent shoppers, and infrequent shoppers. The research methodology includes the following steps: data collection of transactions and customer preferences, data preprocessing, algorithm implementation, and validation and testing. The clustering parameters are based on the number of purchases, number of transactions, and quantity of items purchased. The functional testing results indicate that the system performs well, as all test scenarios were successfully executed. Furthermore, the evaluation using the Silhouette Coefficient (SC) method produced a strong structure status, with an average SC value of 0.97. This result demonstrates that the dataset is highly robust and suitable to serve as a reference model for customer segmentation.]]>
Copyrights © 2025
