Genetic Algorithm Based Bayesian Network for Customers´ Behavior Analysis

In this paper, we propose a credit scoring algorithm to predict seceders of a large Japanese company. We use features based on attributes of customer and genetic algorithm based Bayesian network to detect the seceders. Experimental results on the datasets demonstrate that this method has superior results compared with other recently proposed algorithms, and shows that the proposed method is efficient to detect the seceders from loyal customers.