Credit Card Fraud Detection Using Machine Learning Models

Credit card fraud detection is a critical challenge in the financial sector, necessitating robust and efficient solutions to protect consumers and institutions from significant losses. This research paper explores the application of machine learning algorithms for detecting credit card fraud, emphasizing the use of models such as Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR), K-Nearest Neighbour (KNN), and Decision Tree (DT). By leveraging a comprehensive dataset generated by Kaggle on the transactions made by European cardholders in 2013, the study evaluates the performance of these algorithms. This dataset is highly imbalanced as it only includes 492 fraudulent transactions of 2,84,807 overall transactions. To overcome this imbalance in dataset, in this research we used Synthetic Minority Over-sampling Technique(SMOTE). The paper presents a comparative analysis of the algorithms’ accuracy, precision, recall, and F1-score, highlighting their strengths and limitations. The findings demonstrate that machine learning models can significantly enhance fraud detection capabilities, offering a dynamic and adaptive approach to identifying fraudulent transactions. This research contributes to the development of more effective and efficient fraud detection systems, ultimately improving financial security and reducing economic losses due to fraud.