The Proposed Model Machine Learning of Predicting Bank Churn Customer

Currently, the Bank is an important financial institution for any country, and it helps circulate cash flow to operate the national economy. To do that, any banking business needs a certain number of customers to take advantage of customer benefits. Customer retention is very important in the banking and financial services industry. The increasing development of the internet and artificial intelligence has changed the way data is managed, analyzed, and processed in Banks, in which machine learning has been integrated into customer data analysis for prediction, risk of customer churn. Current methods mainly only use temporal information (such as marital status, gender), have little updated information, and do not take full advantage of the information available at the Bank. In this paper, we propose a model to predict customer churn rate using debit account transaction history data (most customers have debit cards to transact with Banks) to build machine learning models. We collected the sample dataset directly from the transaction information collection system of a commercial Bank in Vietnam (customer Casa data). Thereby using the method of identifying customer churn and machine learning algorithms to build a model to predict customer churn from the Bank, then apply to build a system so that the Bank can come up with strategies to reduce customer churn.