Customer Segmentation Using K-Means Clustering Algorithm and RFM Model

Abstract

The key points in customer segmentation are determining target customer groups and satisfying their needs. Recency-Frequency-Monetary (RFM) analysis and K-Means clustering algorithm are the popular methods for customer segmentation when analyzing customer behavior. In our study, we adapt the K-means clustering algorithm to RFM model by extracting features that represent RFM aspects of home appliances. Customers with similar RFM-oriented features are assigned to the same clusters, while customers with non-similar RFM-oriented features are assigned to different clusters. In the experiments, clustering achieved the determined threshold for Silhouette Score. The resulting clusters were ranked and named by Customer Lifetime Value (CLV) metric, which measures how valuable a customer is to the business.

Kajian untuk Segmentasi Customer Bank dengan Algoritma K-Means

“…Algoritma K-Means, sebagai metode pengelompokan, menawarkan pendekatan yang kuat untuk mengungkap pola yang mungkin tidak terlihat dalam data pelanggan (ASLANTAŞ et al, 2023). Melalui analisis frekuensi transaksi, saldo akun, jenis produk, dan berbagai indikator aktivitas perbankan, kita dapat membedakan karakteristik unik dari setiap segmen pelanggan.…”