A Comparative Analysis of Credit Card Detection Models

Data mining and machine learning are gaining popularity for fraud detection due to their effective results to cater the exponentially growing card transactions that comes with the fast-growing frauds. The aim of this paper is to explore which of the many techniques are capable to detect fraudulent transactions the best. Methods such as logistic regression, decision tree, support vector machine (SVM), Naive Bayes, and random forest are evaluated on their performance based on factors such as accuracy, precision, recall, and F1-score. The results showed that the random forest performed the best among the five methods investigated for credit card fraud detection.