Credit Card Fraud Detection

Abstract

Nowadays, detecting credit card fraud is a major social issue. Credit card usage on e-commerce and banking websites has quickly expanded in recent years. The usage of credit cards in online transactions has made it simple, but it has also increased the frequency of fraud transactions. Modernization will have both beneficial and negative effects. It is always encouraged for banks and e-commerce sites to have automatic fraud detection systems as part of the operations taking place. Huge financial losses could be the outcome of credit card theft. Machine learning approaches offer good answers when searching for ways to stop credit card fraud from happening. When compared to other algorithms currently being used, the proposed system achieves greater accuracy by using a random forest application to solve the issue. All of the fundamental classifiers have the same weight, but the random forest algorithm has a relatively high weight while the others have relatively low weights due to the fact that the bootstrap sampling of decision-making and attribute selection cannot be guaranteed to be equally stable across all classifiers.