Customer Churn Prediction for Customer Relationship Management in the Banking Sector

Machine learning techniques, such is data mining, are used in a variety of fields where a lot of data could be collected and analysed. One of the most popular use cases of data mining is churn modelling and prediction. Predicting churn is used for preventive measures to those, who are most likely to stop using company products, with the goal of retaining these customers. The goal of the project is to construct such model for a CRM department in HCF bank, using Python programming language with the help of open-source libraries. The model is then used for obtaining the list of customers, who are predicted to leave the company. Targeted communications with the aim of retention are then applied to these customers. This paper contains the description of building the model from obtaining the necessary data for training to final results evaluation. The developed solution is expected to outperform the existing solution used in bank in terms of cost and performance.