Customer Churn Prediction by Utilizing Bald Eagle Hunting Strategy with White Tailed Eagle Algorithm and Support Vector Machine Classifier

In the banking industry, customer churn occurs when customers stop using the services provided, leading to a loss of clientele, which negatively impacts the bank's reputation. To retain customers, understanding and analyzing the causes of churn, as well as its predicting is crucial. Machine Learning (ML) algorithms are introduced to analyze and predict customer churn, helping banks identify and retain valuable clients. Classification techniques are employed to classify customer behavior, determining whether a customer will stay or churn. The Bald Eagle hunting strategy-White-tailed eagle-Support Vector Machine (BE-WE-SVM) focuses on predicting customer churn based on the Bank Customer Churn Prediction dataset, using features that enhance churn prediction accuracy. The proposed BE-WE-SVM model was evaluated using existing performance metrics and achieved an accuracy of 99.04%, precision of 98.30%, F1-score of 98.59%, and recall of 98.98%. For the Bank Customer Churn Prediction dataset, the model obtained an accuracy of 99.04%, precision of 98.98%, F1-score of 98.59%, and recall of 98.98%. The proposed method outperforms existing methods, achieving higher accuracy in customer churn prediction.