<![CDATA[This study aims to segment customers to measure their level of loyalty using the RFM (Recency, Frequency, Monetary) model approach combined with the k-Means clustering algorithm. The dataset used comes from the Kaggle site and contains motor vehicle sales data, both cars and motorbikes, with a total of 2,747 transactions. The RFM method is used to calculate three important indicators of customer behavior, namely the last time to make a purchase (recency), purchase frequency (frequency), and total transaction value (monetary). The data is then normalized and grouped using the k-Means algorithm. Based on the results of the Elbow Method and Silhouette Score tests, the optimal number of clusters obtained is four. The segmentation results show four groups of customers with different characteristics, ranging from very loyal customers with high frequency and large transaction values, to customers who have been inactive for a long time. This segmentation is very useful for companies to design more targeted marketing strategies and increase customer retention. This study shows that the combination of RFM and k-Means clustering is able to provide significant insights in understanding consumer behavior and supporting data-based strategic decision making.]]>
Copyrights © 2025
