Customer Relationship Management Using Machine Learning: The Case Of Bunna Bank S.C.
No Thumbnail Available
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Customer Relationship Management (CRM) is critical for enhancing customer satisfaction, loyalty, and profitability in the banking sector. Traditional CRM systems, however, often lack the capability to effectively analyze big volumes of customer data and predict customer behavior. The goal of this research is to create a predictive model for customer classification using machine learning, into high, medium, and low-value segments using Bunna Bank S.C.’s customer dataset to improve CRM practices. In an attempt to develop a customer level prediction model, a total of 64,638 datasets by eleven attributes have been used. The overall accuracy of the model has been
used as the evaluation metric for this study to identify the best classifier. Accordingly, supervised machine learning algorithms including Logistic Regression, Random Forest, Support Vector Machine, K-Nearest Neighbors, and Deep Neural Networks were applied to evaluate their effectiveness in predicting customer levels. Data preprocessing methods such as cleaning, feature selection, and balancing were employed to enhance model performance. The models were evaluated using standard evaluation methods, including accuracy, precision, recall, F-measure, and confusion matrix. The best performing from the selected classifier is Deep Neural Network (DNN) with an accuracy of almost 79.7%, a precision of 70.9%, and recalls also 79.8%; then Random Forest (RF) with an accuracy of almost 76.4%, a precision of 70.2%, and recall also 76.4%; ; Logistic Regression (LR), with an accuracy of almost 62.1%, a precision of 51.1%, and recall 60.3%; Support Vector Machine (SVM) with an accuracy of almost 60.5 %, a precision of 51.2%, and recall also 60.8%; and K-Nearest Neighbour (KNN) with an accuracy of almost 51.0%, a precision of 35.2%, and recall 38.3% respectively. The study concludes that machine learning-based CRM significantly improves Customer Relationship Management, enabling Bunna Bank to deliver personalized services, optimize resource allocation, and strengthen customer retention strategies. Furthermore, the adoption of advanced CRM systems can enhance profitability, support financial inclusion, and modernize Ethiopia’s banking sector.