Hybrid Machine Learning Model for Credit Card Fraud Detection Using SVM, Decision Tree, and Naive Bayes

Pavithra S, Akhila S Babu

Abstract: Credit card fraud is a major challenge in digital financial transactions due to the increasing use of online payments and electronic banking. Traditional fraud detection methods face difficulties in identifying evolving fraudulent patterns while minimizing false alarms. This paper proposes a hybrid machine learning model combining Support Vector Machine (SVM), Decision Tree (DT), and Naive Bayes (NB) using an Adaptive Weighted Voting mechanism. The Credit Card Fraud Detection Dataset is preprocessed using feature standardization and Synthetic Minority Over-sampling Technique (SMOTE) to address class imbalance. The three classifiers are trained independently and their predictions are combined using validation-based weights. Experimental results show that the proposed model achieves 99.08% accuracy, 98.72% precision, 98.15% recall, 98.43% F1-score, and 99.41% ROC-AUC, outperforming the individual classifiers. The results demonstrate the effectiveness of adaptive ensemble learning for reliable credit card fraud detection.

Keywords: Credit Card Fraud, Machine Learning, Adaptive Voting, Class Imbalance, Fraud Detection

How to Cite?: Pavithra S, Akhila S Babu, "Hybrid Machine Learning Model for Credit Card Fraud Detection Using SVM, Decision Tree, and Naive Bayes", Volume 15 Issue 8, August 2026, International Journal of Science and Research (IJSR), Pages: 1857-1858, https://www.ijsr.net/getabstract.php?paperid=SR26825105753, DOI: https://dx.doi.org/10.21275/SR26825105753

Download Citation: APA | MLA | BibTeX | EndNote | RefMan