# Fraud Analytics: A Decade of Research -- Organizing Challenges and Solutions in the Field

**Authors:** Christopher Bockel-Rickermann, Tim Verdonck, Wouter Verbeke  
**arXiv:** [2212.04329](https://arxiv.org/abs/2212.04329)  
**Date:** 2022-12-07  
**Summary page:** https://www.summarizepaper.com/en/paper/2212.04329v1/

## Key points

- Fraud analytics and fraud detection have grown significantly in the past decade, leading to a lack of organization within the field.
- The paper aims to provide an overview of fraud analytics research and establish a more structured framework for the field.
- The authors analyzed nearly 300 records on fraud analytics published between 2011 and 2020, identifying domains of application, challenges faced, performance metrics, and methods used.
- Academic research in fraud analytics covers various objectives, from detecting fraudulent credit card payments to identifying illegitimate insurance claims.
- The authors propose a framework for fraud analytical methods and suggest a keywording strategy for future research to improve organization within the field.
- Accessing suitable public datasets is a major challenge in fraud analytics research.
- The authors outline eight requirements for appropriate data sets based on their findings.
- They structure their sample of literature into an online database accessible to fellow researchers for investigation and potential expansion.

## Summary

The field of fraud analytics and fraud detection has experienced significant growth in the past decade, resulting in a wide range of research topics and a lack of organization within the discipline. This paper, authored by Christopher Bockel-Rickermann, Tim Verdonck, and Wouter Verbeke, aims to provide an overview of fraud analytics research and establish a more structured framework for the field. The authors analyze nearly 300 records on fraud analytics published between 2011 and 2020, systematically identifying the primary domains of application, challenges faced, performance metrics, and methods used. The focus of academic research in fraud analytics spans from detecting fraudulent credit card payments to identifying illegitimate insurance claims. With such diverse objectives, there is a need for greater organization within the field. To address this issue, the authors propose a framework for fraud analytical methods and suggest a keywording strategy for future research. One major challenge in fraud analytics is accessing public datasets that are suitable for research purposes. In order to assist the research community further, the authors outline eight requirements for appropriate data sets based on their findings. Additionally, they structure their sample of literature into an online database that is accessible to fellow researchers for investigation and potential expansion. Overall, this paper provides valuable insights into the current state of fraud analytics research while offering solutions to improve organization within the field. By analyzing existing literature and proposing frameworks and strategies for future studies, it contributes to advancing knowledge in this important area of study.