article · Graduate Journal of Science and Technology
This study investigated customer churn prediction in the banking sector using data from Kaggle. It evaluated five machine learning models: Logistic Regression, Random Forest, Gradient Boosting, Artificial Neural Network, and Stacking Classifier. The models were assessed based on Accuracy, Precision, Recall, Specificity, and Area Under the Curve (AUC). The research found that ensemble learning methods performed best, with the Stacking Classifier achieving the highest accuracy of 0.8733. Gradient Boosting demonstrated the best balance between precision and recall, and the highest AUC of 0.8820. These findings highlight the effectiveness of ensemble learning in improving predictive performance for customer retention.
Understanding and predicting customer churn is crucial for banks to retain their customers and maintain profitability. By identifying customers at risk of leaving, banks can implement targeted strategies to improve satisfaction and loyalty, which is vital in a competitive market.
This research provides applied insights for financial institutions seeking to minimise customer churn. The evaluated machine learning models, particularly ensemble methods, could be integrated into existing customer relationship management systems to identify at-risk customers. This would enable banks to deploy proactive retention strategies, moving towards real-world application in customer service and marketing departments.
AI-generated from the published abstract. Always read the original work before citing.
This study investigates the prediction of the phenomena of customer churn issue in the banking using the bank data downloaded from Kaggle. Customer churn prediction is a critical aspect of customer relationship management, allowing businesses to implement proactive retention strategies. Study evaluates the performance of five machine learning models, Logistic Regression (LR), Random Forest (RF), Gradient Boosting (GB), Artificial Neural Network (ANN), and Stacking Classifier in predicting customer churn. The models were assessed based on the five key metrics: Accuracy, Precision, Recall, Specificity, and Area Under the Curve (AUC). The results indicate that ensemble learning, specifically the Stacking Classifier, outperforms individual models by achieving the highest accuracy (0.8733). Gradient Boosting demonstrated the best balance between precision and recall, with the highest AUC (0.8820). The findings highlight the effectiveness of ensemble learning in improving predictive performance, offering valuable insights for businesses seeking to minimize churn rates in accordance of high competition.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.33914/gjst.v21i2.407
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
