• Krishna Gupta

The number of banking transactions increased as a consequence of the widespread use of credit cards as a payment mechanism due to the proliferation of e-commerce services and other technological advancements. Customers and the bank alike suffer enormous financial and reputational harm as a result of fraudulent conduct made possible by security holes in our banking systems. An estimated large sum of money is lost financially each year as a consequence of financial fraud in banks. Early discovery aids in the mitigation of fraud by allowing for the development of a countermeasure and the recovery of such losses. This research proposes a machine learning-based method to effectively aid in fraud detection. Comparing six popular ML algorithms—AdaBoost, NBC, SVM, KNN, ANN, and XGBClassifier—aims to identify the best model for fraud detection. Moreover, we addressed the problem of class imbalance by using the SMOTE. A 99.88% accuracy rate for fraud detection made the XGBoost Classifier the most effective classifier. This study demonstrates the remarkable accuracy score of the XGBoost Classifier in identifying financial fraud on balanced datasets. The information and conclusions offered here make a significant addition to the continuing work being done to create efficient fraud detection systems for a banking sector.

Fraud Identification, Banking Transactions, credit card frauds, machine learning, XGBoost

"A Fraud Identification in Banking Transactions with Machine Learning Technique in Financial Management ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 8, page no.e387-e394, August-2024, Available :http://www.jetir.org/papers/JETIR2408436.pdf

"A Fraud Identification in Banking Transactions with Machine Learning Technique in Financial Management ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 8, page no. ppe387-e394, August-2024, Available at : http://www.jetir.org/papers/JETIR2408436.pdf