|
Evaluating Machine Learning Methods for Solving Class Imbalance in Banking Customer Data: A Comparative Study |
|---|---|
| รหัสดีโอไอ | |
| Creator | Benjawan Rodjanadid |
| Title | Evaluating Machine Learning Methods for Solving Class Imbalance in Banking Customer Data: A Comparative Study |
| Contributor | Jirakit Boonmunewai, Teerapat Chantaraksa |
| Publisher | KKU Science Journal |
| Publication Year | 2567 |
| Journal Title | KKU Science Journal |
| Journal Vol. | 52 |
| Journal No. | 3 |
| Page no. | 349 - 362 |
| Keyword | Churn Prediction, Synthetic Minority Over-sampling Technique, Decision Tree Classifier, Naïve Bayes Classifier, Support Vector Machine |
| URL Website | https://ph01.tci-thaijo.org/index.php/KKUSciJ/article/view/258592 |
| Website title | Thai Journal Online (ThaiJO) |
| ISSN | 3027-6667 |
| Abstract | The goal of this research was to address an imbalance problem that affects churn prediction for bank customers. In this study, we examined two sampling techniques Synthetic Minority Over-sampling Technique (SMOTE) and random under-sampling along with three predictive models: the decision tree classifier, Naïve Bayes classifier, and support vector machine classifier. The results indicated that the support vector machine classifier, when combined with SMOTE, was the most effective, achieving a recall of 92.99%, an F-score of 91.37%, an area under the curve (AUC) of 96.4%, and a false negative rate of 7.01%. |
