Research on port customer classification based on RBF Neural Network

Abstract

Customer is an important strategic resources for enterprise, only use effective customer management can promote the development of the enterprise. Customer classification management is the key to customer relationship management, and the key to effective customer classification is to select the appropriate classification index and effective classification method. In this paper, the customer classification index is extract from the analysis of customer value, at the same time, customer classification method is selected by comparing the different classification methods. Based on the comprehensive analysis of Subtraction Clustering, fuzzy K-Prototypes algorithm and the improved Particle Swarm Optimization (PSO) algorithm, the algorithm of RBF Neural Network was improved and a new RBF Neural Network model was built. Then I apply the model to the port on the customer classification. Finally using MATLAB to simulate and verify the validity of the model on the customer classification.