Automating Credit Card Limit Adjustments Using Machine Learning

TL;DR

This paper presents a machine learning approach to automate credit card limit adjustments in Venezuelan banks, replacing manual committee decisions with models that balance accuracy, cost, and interpretability, achieving high agreement with human decisions.

Contribution

It introduces a cost-sensitive neural network and XGBoost models for automating credit limit decisions, demonstrating superior performance and agreement with manual decisions.

Findings

Models achieved near-perfect agreement with committee decisions.

XGBoost and neural networks outperformed baseline methods.

Cost-sensitive learning improved decision accuracy and fairness.

Abstract

Venezuelan banks have historically made credit card limit adjustment decisions manually through committees. However, since the number of credit card holders in Venezuela is expected to increase in the upcoming months due to economic improvements, manual decisions are starting to become unfeasible. In this project, a machine learning model that uses cost-sensitive learning is proposed to automate the task of handing out credit card limit increases. To accomplish this, several neural network and XGBoost models are trained and compared, leveraging Venezolano de Credito's data and using grid search with 10-fold cross-validation. The proposed model is ultimately chosen due to its superior balance of accuracy, cost-effectiveness, and interpretability. The model's performance is evaluated against the committee's decisions using Cohen's kappa coefficient, showing an almost perfect agreement.