A Classifier Ensemble Model Based on GMDH-Type Neural Network for Customer Targeting

With the speedy development of information technology, database marketing has become a hot topic for both marketing practitioners and scholars, and customer targeting modeling is a top priority in database marketing. To overcome the limitations of the existing models, this study proposes a classifier ensemble model based on group method of data handling (GMDH) type neural network for customer targeting. It first utilizes GMDH-type neural network to select the key explanatory variables, trains a series of base classification models, and then conducts the classifier ensemble selection based on the forecasting results of the base models by GMDH-type neural network again to get the final ensemble model. The empirical analysis results show that the hit rate of the proposed model is better than that of some existing models, and it can bring more profits for the enterprise.