Customer segmentation model based on two-step optimization in big data era

摘要

With the advent of the era of big datasets, real-time data is becoming increasingly important in assisting the decision making process for commercial banks. In this paper, we develop a two-step optimization model (FSGA-FCEN) based on genetic algorithm (GA) and cluster ensemble (CE) for customer segmentation. Firstly, the key attributes are selected using GA. Then FCEN algorithm is used to segment customers into small groups. Taking 3544 customers in a commercial bank as samples, empirical results show that, compared with K-means, FCM and MAJ models, two-step model is an efficient and practical tool for customer segmentation.