Credit Card Fraud Detection using Imbalanced Learning

This paper proposes a credit card fraud detection system that addresses the challenge of
highly imbalanced transaction data. Since fraudulent transactions represent only a small fraction of
total transactions, traditional models tend to favour the majority class. To overcome this issue, under
sampling and SMOTE were applied to balance the dataset. Machine learning algorithms such as
Logistic Regression and SVM were used for classification. The results demonstrate improved fraud
detection performance, particularly in terms of recall, while maintaining a reasonable balance between
precision and accuracy.
Keywords: SMOTE, SVM, Logistic Regression.