Customer behavior-based fraud detection of credit card using a random forest algorithm

Abstract

Credit card use has become necessary due to the rapid growth of e-commerce and the Internet. Because of the growing use of credit cards, the number of scams related to them has also grown. Such issues may be addressed through data science, which, when combined with machine learning, cannot be underestimated. This goal, “Credit Card Fraud Detection,” aims to uncover the structure of a data set using ML (machine learning). There are a variety of strategies that may be used to identify fraudulent activities. The primary objectives of this approach are to achieve the highest possible degree of precision, a high rate of successfully detecting fraudulent activity, and a low number of false positives. Customer behaviors have been included in this proposed work to identify fraudulent activities. The Random Forest Algorithm has the highest accuracy and MCC scores of all the algorithms. It has been found that the random forest algorithm has the greatest accuracy (94.4 percent) in detecting fraudulent credit card activity. Kaggle provided the dataset that was used in the analysis of credit card fraud