Document Type : Original Research Paper
Authors
Department of Information Technology, Faculty of Industrial Engineering, Khwaja Nasiruddin Toosi University of Technology, Tehran, Iran
Abstract
With the saturation of markets, organizations have realized that keeping customers, especially valuable customers, should be at the center of their management strategies because the cost of attracting new customers is higher than the cost of keeping existing customers. The insurance industry is no exception to this and due to the low switching cost, it faces customers who want to change their current company. The current research examines the effect of the variables of relationship length, purchase delay, purchase frequency, financial value, profitability and the group of purchased products in the valuation of the studied customers. For this purpose, a questionnaire survey tool was used in order to know the opinions of experts regarding the variables affecting the customers' valuation. The results show that the variables of frequency of purchase, duration of cooperation and the number of insurance groups purchased are the most important from the point of view of experts in the valuation of customers. Then, by extracting important variables and factors on the reversion of insurance policy holders, the influence and importance of these factors on the reversion of valuable customers has been investigated, and further, using the variables identified in the previous step, the development of the predictive model of reversion has been done. With different models (neural network, decision tree, support vector machine, and logistic regression) prediction modeling and the accuracy of the built models have been evaluated. The results show that the C5.0 decision tree model has a higher accuracy and accuracy in predicting diversion than other models.
Keywords
Letters to Editor
Send comment about this article