Market segmentation is one of the most fundamental concepts in market analy- sis. The interpretation of the quality of the segments is still a crucial task. To solve this prob- lem, market segmentation uses cluster analysis as a very useful tool to segment customers and uses clustering evaluation criteria to extract the optimal number of segments. In the literature, different clustering methods were used in Market segmentation like k-means, Self-organizing map, etc. In this paper, we study the Finite Mixture Models, or model based clustering and its good performance in market segmentation along with the multi-SOM clustering method. Then, we apply both of them on a Tunisian banking data set using different evaluation criteria. Re- sults show the efficiency of both approaches in determining the optimal number of clusters compared with k-means method. Keywords:
CITATION STYLE
Wagner, R., Scholz, S. W., & Decker, R. (2005). The Number of Clusters in Market Segmentation. In Data Analysis and Decision Support (pp. 157–176). Springer-Verlag. https://doi.org/10.1007/3-540-28397-8_19
Mendeley helps you to discover research relevant for your work.