Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{202640,
        author = {Ashish Ravindra Mewal and Dr. Syed Sumera Ali and A.G.Gaikwad and A.T. Jadhav and Dr. D.L. Bhuyar and Dr.G.B.Dongre},
        title = {A Comparative Analysis of Synthetic Minority Over-Sampling Technique (Smote) In Enhancing Credit Card Fraud Detection Systems},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {8684-8690},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=202640},
        abstract = {The rapid growth of digital payment systems has made credit card fraud a major concern, necessitating reliable and efficient fraud detection systems. Machine learning (ML)models have demonstrated significant potential in detection of fraudulent transactions, but are often limited by the extreme class imbalance of credit card data, with fraudulent transactions only a small fraction of all transactions. This paper entails a comparative evaluation of the Synthetic Minority Over-Sampling Technique (SMOTE) in credit card frauds detection systems improvement. Several operational ML models are evaluated under two experimental settings: one with and one without SMOTE. These models include Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Gradient Boosting (GB), among others. Imbalance-sensitive measures of performance are used to quantify performance i.e., recall, F1-score and AUC-ROC. The findings illustrate that models that are trained without SMOTE are quite accurate but cannot be used to prevent detection of fraudulent transactions because of poor recall. By contrast, the use of SMOTE substantially enhances recall and F1-score on all models, and the greatest improvement in performance is observed in ensemble classifiers. The results confirm that SMOTE is a good tool in reducing the effects of class imbalance, false negatives, and increasing the usefulness of machine-learned credit card fraud detection systems in practice.},
        keywords = {Credit Card Fraud Detection, SMOTE, Class Imbalance, ML, Oversampling Techniques, Fraud Analytics},
        month = {May},
        }