Fraud continues to be an important topic in the financial world helping detect when funds are mismanaged.

Nilson reports that U.S. card fraud (credit, debt, etc) was reportedly $9 billion in 2016 and expected to increase to $12 billion by 2020. Due to these nefarious acts; it is in the best interest of any organization to ensure the work of most customers. Thus it is appropriate to respond to this fraud.

Hence there is much written about these methods to detect fraud in order to provide mitigation to an organization.

Introduction

Some of the traditional ways to catch fraud are as follows Anomaly Testing

Reference

https://blogs.oracle.com/datascience/fraud-detection-using-autoencoders-in-keras-with-a-tensorflow-backend