Credit Card Fraud Detection System Using Machine Learning

Credit Card Fraud Detection System Using Machine Learning

Alok Kumar Srivastava1, Anjali Rao2,Ankita Sahani3,Shikha Rao4

1Assistant Professor, Department of Computer Science & Engineering, Buddha Institute of Technology, Gorakhpur, India

2,3,4B.Tech Student, Department of Information Technology, Buddha Institute of Technology, Gorakhpur, India

Email- alok6123@gmail.com, anjalirao5apr@gmail.com, ankitasahani329@gmail.com, shikharao23oct@gmail.com

Abstract

Events involving credit card fraud happen regularly and end up costing a lot of money. The number ofson line transactions have grown in large quantities and online. A significant portion of these transactions are credit card transactions. As a result, banks and other financial organizations provide very valuable and in-demand credit card fraud detection programs. Fraudulent transactions can take many different forms and fall under several categories. The four primary fraud incidents in real-world transactions are the topic of this research. A variety of machine learning models are used to address each fraud and the optimal approach is ultimately chosen after examination. This assessment offers a thorough guidance for choosing the best algorithm with regard to the type of frauds and with the use of an appropriate performance measure, we illustrate the evaluation. Real-time credit card fraud detection is another crucial topic that we cover in our project. To determine if a specific transaction is legitimate or fraudulent, we use predictive analytics performed by the implemented machine learning models and an API module. We also evaluate a cutting-edge approach that successfully tackles the skewed distribution of data. According to a private disclosure agreement, the financial institution provides the data for our studies.

Keywords- Skewed distribution, real-time credit card fraud detection, confidential disclosure agreement, fraud detection system, and credit card frauds.