Managing the Imbalanced Data of Cross-sell Strategy for Credit Card Holders

Abstract

Cross-selling insurance products, which is part of banking activities, has grown in importance in today's banking industry. Because of the negative customer responses, highly imbalanced data is taken into account. The study introduces the concept of management imbalanced data and a forecasting model for predicting customer satisfaction response rates, including with specific data. In this study, a simulated dataset was created that replicated the statistical properties of the original data. The data is then adjusted using Under Sampling, Over Sampling, and the Synthetic Minority Over-sampling Technique (SMOTE). In addition, an algorithm is used to select the precise factor. Furthermore, regression analysis is used to forecast the effectiveness of data analysis. This study demonstrates that the Synthetic Minority Over-sampling Technique (SMOTE) is an effective method for improving the imbalanced cross-selling insurance data.