Research on bank customer satisfaction classification on random forest algorithm

As the customer-orientated business philosophy continues to deepen the impact on our banks, customer satisfaction has more and more important implications for bank’s profitability and development. How to efficiently evaluate bank customer satisfactionhas become an urgent issue to be solved. It is of great significance to establish an accurate customer satisfaction prediction model for commercial banks. In this paper, we use Random Forest algorithm based on Grid Search CV to establish a two-classification prediction model for bank customer satisfaction and compare Random Forest algorithm with SVM as well as Logistic Regression algorithms in terms of prediction accuracy for bank customer classification. In order to providesome effective reference and basis for banks to select targeted customers and improve service levels to attain more profit.