Title: Fraud and anomaly detection models in banks: a systematic analysis and literature connection

Authors: Alex Cerqueira Pinto; Mathias Schneid Tessmann; Alexandre Vasconcelos Lima

Addresses: Bank of Brazil, Brasília, Brazil; University of Brasília, Brasília, Brazil ' Brazilian Institute of Education, Development and Research (IDP), Brasília, Brazil ' Brazilian Institute of Education, Development and Research (IDP), Brasília, Brazil

Abstract: This paper seeks to analyse and verify existing connections in the literature on fraud detection in banks. For this, 227 papers published until September 2022 in the Web of Knowledge through the PRISMA protocol are analysed and classified. The works were identified using the keywords 'fraud', 'model', 'detection', 'banking' and 'risk' and classified into 12 categories, such as type of study, approach, cut, design, nature, the purpose of study, method, spatial scope, period of study, focus, data used and results. Based on the classification, statistics of complex networks are also used to identify the existing citation connections between them. The results show that there is a dissemination of the use of machine learning techniques together with business rules to detect possible cases of fraud and a growing increase in cases of fraud with social engineering. These findings are useful for the scientific literature that investigates operational risk professionals of banks.

Keywords: fraud detection; anomaly detection; systematic literature review; bibliometric; machine learning; banks.

DOI: 10.1504/IJBBM.2024.140372

International Journal of Bibliometrics in Business and Management, 2024 Vol.3 No.2, pp.182 - 205

Received: 16 Jun 2023
Accepted: 29 Nov 2023
Published online: 05 Aug 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article