Customer Value Classification Model Based on Improved TOPSIS Method and BP-Adaboost Algorithm

Identifying customer groups with different consumption values can help enterprises understand and analyze customers' purchasing behaviors and preferences, so as to effectively improve their operating profits. In this paper, the customer value evaluation index system is constructed on the basis of existing research on customer value classification using entropy weight method and improved TOPSIS method, and different grades are classified according to the comprehensive score of customers. Subsequently, the customer comprehensive score is used as the output layer, and the BP-Adaboost algorithm is used to construct the customer value classification model. The empirical results show that the improved TOPSIS method is more advantageous in terms of error control and classification accuracy, and the BP-Adaboost algorithm outperforms the traditional BP neural network in the customer value classification model, which improves the classification precision and accuracy. This study provides a powerful tool for dealing with the customer value classification problem, which helps enterprises make more accurate decisions in customer relationship management and resource optimization allocation.