Table of Content

Open Access iconOpen Access


A Novel Cardholder Behavior Model for Detecting Credit Card Fraud

Yiğit Kültür, Mehmet Ufuk Çağlayan

Computer Engineering Department, Boğaziçi University, Istanbul, Turkey

* Corresponding Author: Yiğit Kültür, email

Intelligent Automation & Soft Computing 2018, 24(4), 807-817.


Because credit card fraud costs the banking sector billions of dollars every year, decreasing the losses incurred from credit card fraud is an important driver for the sector and end-users. In this paper, we focus on analyzing cardholder spending behavior and propose a novel cardholder behavior model for detecting credit card fraud. The model is called the Cardholder Behavior Model (CBM). Two focus points are proposed and evaluated for CBMs. The first focus point is building the behavior model using single-card transactions versus multi-card transactions. As the second focus point, we introduce holiday seasons as spending periods that are different from the rest of the year. The CBM is fine-tuned by using a real credit card transaction data-set from a leading bank in Turkey, and the credit card fraud detection accuracy is evaluated with respect to the abovementioned two focus points.


Cite This Article

Y. Kültür and M. U. Çağlayan, "A novel cardholder behavior model for detecting credit card fraud," Intelligent Automation & Soft Computing, vol. 24, no.4, pp. 807–817, 2018.

cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1270


  • 809


  • 0


Share Link