An Intelligent Method for Credit Card Fraud Detection Using Data Mining Techniques

In recent years, billions of dollars in losses have been caused by fraudulent credit card transactions, representing a serious and growing problem. To mitigate the damage from such transactions, data mining techniques are widely employed for credit card fraud detection, utilizing approaches such as classification and clustering with machine learning algorithms. This study focuses on supervised learning, in which machine learning algorithms are trained on labeled datasets to construct predictive models. A major challenge in this domain is the highly imbalanced class distribution within the datasets, as the number of fraudulent transactions is significantly lower than that of legitimate transactions. This paper examines strategies for addressing imbalanced data in machine learning algorithms and proposes an optimized method for detecting and identifying fraud on both original and balanced datasets. In this study, the performance of classification techniques is compared using well-known methods, including C۵.۰ decision trees, Support Vector Machines (SVM) with Sigmoid, Linear, and RBF kernels, and neural networks.

کلیدواژه ها:

Electronic banking ، Data mining ، Credit card fraud ، Imbalanced Data