Previous Issue

Asian Journal of Information and Communications 9(2)
Letter from editors Special section "Recent Progress in ICT Application in China" General sect....

Call for Paper

Add to Bookmark

Editorial Office

Asian Journal of Information and Communications

Customer analysis of the securities companies based on modified RFM model and simulation by SOM neural network Juan Cheng, Xiongwei Zhou and Bixuan Fu

  • Page:58-66
  • 2018-05-02

 The customers of the securities company have been classified according to the balance of their accounts for a long time. The method cannot measure the value of customers dynamically. Some securities companies apply RFM model to classify customers in recent years. But only recency, frequency and monetary cannot evaluate customers of the securities companies well. The commission of the trading varieties should be taken into account when calculating recency, frequency and monetary as it will affect the value of customers. This paper proposes a modified RFM model in order to precisely evaluate the customers of the securities companies. Every variety is weighted according to the commission in the modified RFM model. In order to evaluate the modified RFM model, the paper classifies 5,000 randomly chosen customers by self-organizing feature map neural network. The results show the customers are classified into eight categories and the number of general customers and valueless customers increase significantly which reflects that the modified RFM model can classify customers strictly and precisely. So, the modified RFM model is helpful for the securities company to develop corresponding service strategies.

Keywords: securities companies, customer analysis, RFM model, SOM neural network, simulation

  PDF Download
  • Facebook
  • Share onTwitter