Abstract
Creditcard fraud detection mechanism is the most needed one in the present world. This is due to the development of e-commerce and online transaction. The main aim of this paper is to predict fraudulent transactions in a credit card transaction data set using statistical methods. Statistical methods based on Logistic Regression and Random Forest are developed and applied to credit card fraud detection problems. This work mainly focuses on machine learning algorithms as a classification model. The classification model is used for categorizing the dataset into fraud and non-fraudulent transactions. This project comprises methods for analyzing the fraud data to extract meaningful statistics and other characteristics of the data. The use of a model is to predict fraud with different statistical methods. This paper is one of the first to compare the performance of Logistic Regression and Random Forest methods in credit card fraud detection with a transaction data set using R. The results of the two algorithms are based on accuracy, precision, recall, and F1-score. The ROC curve is plotted based on the confusion matrix. The Random Forest and the Logistic Regression algorithms are compared and the algorithm that has the greatest accuracy, precision, recall, and F1 score is considered the best algorithm that is used to detect fraud.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have