Enhanced Technique for Credit Card Extortion Detection Using Extreme Gradient Boosting Algorithm

Abstract

The most common issue in the modern world is the identification of credit card fraud. This is a result of the expansion of both online commerce platforms and online transactions. In utmost cases, credit card fraud occurs when the card is stolen and used for any unauthorised exertion, or indeed when the fraudster utilises the card's information for their own gain. The credit card scam detection system was introduced with machine learning algorithms to catch these actions. Financial fraud is a growing problem in the financial industry with long-term consequences. It becomes difficult for two main reasons: first, the profiles of legitimate and fraudulent behaviour are always changing, and second, the data sets for credit card fraud are quite biased. The main objectives of this study are to identify the various types of fraudulent credit cards and to investigate alternate fraud detection techniques. On severely skewed credit card fraud data, it evaluates the performance of Decision tree, Random Forest, Logistic Regression and Extreme Gradient Boosting (XG Boost).