Customer segmentation and visualization by combination of self-organizing map and cluster analysis

Abstract

To perceive the characteristics of customers from market information, it is necessary to aggregate the market information. So, we used a self-organizing map (SOM), arecency, frequency and monetary (RFM) analysis, and other methods to propose the classification of customers. In addition, we propose a visualization of the analysis results. In this study, we consider the combination of cluster analysis and the SOM for further consolidation of small clusters and visualization.