New approach clustering algorithm for customer segmentation in automobile retailer industry

Abstract:This research presents a new method for automatic customer segmentation based on usage behavior, without using a human specialist. In addition, the research focuses on profiling customers and finding a relation between the profile and the segments. The customer segments were constructed by applying modified Gustafson-Kessel clustering algorithm. The customer’s profile was based on personal information of the customers. A novel Support Vector Machines was used to estimate the segment of a customer based on his profile.To demonstrate the efficiency of the proposed method, this work performs an empirical study of a Nissan automobile retailer to segment over 5000 customers. This led to solutions for the customer segmentation with six segments.