A Novel Machine learning Algorithms used to Detect Credit Card Fraud Transactions

Abstract

During the Covid-19 pandemic, the world was under lockdown, and everyone was inside their home. There are so many restrictions for going out, so many companies introduced online shopping, and this online shopping helped more people; the e-commerce platform also increased their revenue; at the same time, online fraud has also risen worldwide. Everyone adopted online shopping during the pandemic. In 2019 India's 2019 credit/debit card fraud rate was 365, according to the National Crime Record Bureau. The developed countries are the highest rate of credit card fraud in 2020 compared to India; for that reason, we have to implement mechanisms that can detect credit theft. The machine learning algorithm with the R program will play an essential role in credit card fraud detection. The following machine learning algorithm will have used for credit card fraud, Random Forest, Logistic regression, Decision trees, and Gradient Boosting Classifiers. The European bank dataset used in our research and the dataset size is 284808. Here we used two classes, the first one is called the positive class (fraud transactions), and the second one is the negative class (genuine transactions). The final result will show us the outperforms of our proposed system.