Research on the Optimization of Bank Precision Marketing Strategies Based on Catboost

The competition in the financial industry is becoming increasingly fierce, and the digital transformation of the banking industry is accelerating. To improve effectiveness of banks and enhance their core competitiveness, the accuracy of bank marketing and the ability to accurately predict customer needs should be enhanced. Therefore, this paper studies the issue of using appropriate machine learning models to predict bank marketing and finds the machine learning model suitable for completing the task of bank marketing prediction. This paper uses a dataset of American bank marketing on Kaggle for research. Through data visualization analysis of the dataset, it is found that the overall data shows a significant nonlinear correlation. To learn the significant nonlinear correlation of the data, five machine learning models, namely KNN, logistic regression model, Catboost, GBDT, and random forest, are designed. By comparing the performance of the five machine learning models, it can be obtained that the recall rate, accuracy rate, and AUC value of the Catboost model are the highest among the five models, and its performance is the best. Therefore, it indicates that the Catboost model is very suitable for such bank marketing prediction tasks. This study deeply explores bank marketing prediction based on Catboost, providing new perspectives and methods for related fields, and also providing new ideas for accelerating the digital transformation of banks, accurately predicting customer needs, and improving marketing benefits.