Efficient Credit Card Fraud Detection Based on Multiple ML Algorithms
Neha Ahirwar, Divakar Singh, Kamini Maheshwar
Abstract
The occurrence of credit card fraud has become a major worry in the current digital era, impacting financial institutions and customers alike. Effective and reliable fraud detection systems are in high demand as fraudulent actions continue to change. Sophisticated machine learning algorithms have become essential instruments in tackling this problem. The design and assessment of a credit card fraud detection system based on well-known machine learning algorithms— Random Forest, Logistic Regression, Decision Tree, and XGBoost—is thoroughly investigated in this study.The increasing number of digital transactions has given criminals additional chances to take advantage of holes in payment systems. As a result, the requirement for proactive and flexible fraud detection systems has increased significantly. In order to solve this urgent issue, this study investigates how well machine learning algorithms can detect fraudulent credit card transactions.The machine learning algorithms that are the subject of this study were selected due to their demonstrated efficacy in a number of fields, including the identification of credit card fraud. The models Random Forest, Logistic Regression, Decision Tree, and XGBoost are highly respected for their capacity to accurately predict complex patterns. A variety of real and fraudulent credit card transaction datasets are used to develop each method and evaluate its performance in a controlled setting.