Customer Clustering for an Online Store Using the K-means Algorithm

Customer Clustering for an Online Store Using the K-means Algorithm

Ehsan Eramian1 Goudarz Saadati Moghaddam2

1) M.Sc. Student in Information Technology Engineering – E-Commerce,
2) Assistant Professor, Department of Information Technology Engineering – E-Commerce, Islamic Revolution University (Jame-e Enqelab),

IOI Code : ABDE-ABKGCB
Publication : 6th International Conference on Advanced Research in Management And Humanities(huconf.com)
Abstract :
    E-commerce systems enable marketing and product introduction through the internet, and traditional marketing is gradually being replaced due to its lower efficiency. To enhance profitability, customer segmentation is performed based on shared characteristics to determine the appropriate approach for each customer. This is carried out using various criteria, including behavioral, geographical, demographic, and psychological factors. In this research, the RFM model, which is recognized as one of the suitable tools for extracting behavioral features and measuring customer value, is employed. Additionally, to more accurately cover behavioral dimensions and measure the average transaction value, the composite indicator Average Monetary Value per Purchase or AMV is used as a complementary component. In the clustering process, the k-means++ algorithm, which is an improved version of k-means, is utilized. To optimize and validate the number of clusters, in addition to the Elbow and Silhouette methods, the Davies–Bouldin index and the Bootstrap ARI statistical test have been used to ensure the stability and reproducibility of the segmentation. The results led to the identification of three distinct customer groups: Strategic, High-Searchers, and Transient Customers. Ultimately, this approach helps online businesses enhance customer satisfaction and loyalty among potential customers by delivering customized experiences.

Keywords : customer segmentation rfm model k means algorithm elbow method silhouette analysis