Credit Card Transaction Fraud Detection

Nowadays with the increase of payment through e-commerce, payment fraud has become an important study for online payments. Many unauthorized people gain new methodologies in order to get access to other people’s financial account in an illegal way. In a way, the modern payment systems have some loopholes in which the reputed organizations, banks, merchants, and the customers are all in high risk. Credit card fraud occurs when there is a breach in website security or due to the carelessness of the card’s owner. Due to these data breaches, credit card data can be used in online transactions illegally by other people who have the data captive with themselves. In our model, we would use random forest algorithm with artificial neural networks. Random forest can help see transactions individually into decision trees. So many transactions have multiple decision trees. Then each of these decision trees helps to predict an outcome; when put together, all predictions in aggregate from all the decision trees give us a more stable and accurate forecast.