PERFORMANCE EVALUATION OF MACHINE LEARNING METHODS FOR CREDIT CARD FRAUD DETECTION USING SMOTE AND ADABOOST

Abstract- In this work, SMOTE (Synthetic Minority Over- sampling Technique) and AdaBoost (Adaptive Boosting) algorithms are used to assess the effectiveness of machine learning techniques for detecting credit card fraud. The dataset employed in this study is very unbalanced, with a much higher proportion of legitimate transactions than fraudulent ones. Six machine learning methods—Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbours, Support Vector Machines, and Artificial Neural Networks—have been tested to determine how well they perform. These algorithms are assessed using a variety of measures, including accuracy, precision, recall, and F1- score. The outcomes demonstrate that the SMOTE technique successfully balances the dataset and enhances the efficiency of each programme. The AdaBoost algorithm also enhances the performance of the Random Forest, Artificial Neural Networks, and Decision Tree algorithms. The study's findings may be useful.This study evaluates the performance of machine learning methods for credit card fraud detection using SMOTE (Synthetic Minority Over- sampling Technique)