A Hybrid Fuzzy Logic and Deep Learning Method for Fraud Detection: Application to Credit Cards

The credit card has become an essential part of business operations, both nationally and globally. It plays a very important role in today's economy. While the use of credit cards offers considerable benefits when used responsibly and with care, considerable financial and credit damage can be caused by fraudulent default. Many techniques have been proposed to combat the growth of credit card defaults. However, all these techniques have the same goal of preventing credit card defaults; each has its own characteristics, advantages, and disadvantages. In this paper, we applied two data mining techniques: Artificial Neural Networks and Fuzzy C-Means to this problem, we propose a hybridization between these two models and also, we show significant results of the three methods on real financial data. Therefore, out of all the methods used, good performances are obtained by using our hybrid model to fraud detection.