ENSEMBLE-BASED FRAUD DETECTION IN NIGERIAN BANKING: A SYNTHETIC DATA BENCHMARK AND COST-SENSITIVE ANALYSIS  GBOLAHAN ADENIRAN IDOWU∗ AND JOSIAH ENDURANCE OWOLABI  ABSTRACT. Fraud detection in Nigerian banking is critically hampered by a scarcity of authen

GBOLAHAN ADENIRAN  IDOWU; JOSIAH ENDURANCE  OWOLABI

GBOLAHAN ADENIRAN IDOWU DEPARTMENT OF MATHEMATICS, LAGOS STATE UNIVERSITY, OJO, LAGOS STATE, NIGERIA.
JOSIAH ENDURANCE OWOLABI DEPARTMENT OF STATISTICS, UNIVERSITY OF LAGOS, AKOKA, LAGOS STATE, NIGERIA.

Keywords: Cost-sensitive Learning, Fraud Detection, Nigerian Banking, Random Forest, SHAP Interpretability, Synthetic dataset, XGBoost

Abstract

Fraud detection in Nigerian banking is critically hampered by a scarcity of authentic transaction data for model development, a challenge exacerbated by the rapid growth of digital payments. To address this foundational data gap, this study introduces a novel, high- fidelity synthetic benchmark dataset of 1,000,000 financial transactions, meticulously calibrated to reflect the fraud patterns reported by the Nigeria Inter-Bank Settlement System (NIBSS). Using this dataset, we develop a comprehensive analytical framework to evaluate the economic efficacy of advanced machine learning models. The results demonstrate a substantial potential for fraud loss reduction: optimized Random Forest models achieved a 69.1% decrease in simulated fraud-related costs (from 81.8M to 25.2M) while maintaining perfect precision. Alternatively, XGBoost delivered superior recall (74.6%) with an F1-score of 0.854, providing a strategic option for institutions prioritizing fraud detection rates. A SHAP analysis identified transaction amount and associated behavioral features as the strongest fraud indicators and highlighted Web and Mobile channels as requiring enhanced monitoring. This paper makes three principal contributions: the first publicly available, NIBSS-calibrated fraud detection dataset for Nigeria, addressing a pivotal data scarcity in African financial research; empirically validated evidence that ensemble methods, combined with threshold optimization, can reduce fraud costs by up to 69%; and actionable implementation guidelines for Nigerian banks operating within existing regulatory compliance frameworks. The synthetic data methodology offers a replicable and privacy- preserving blueprint for other emerging markets facing similar constraints on data access and availability.

References

[1] Nigeria Inter-Bank Settlement System. Electronic Payment Transactions Statistics: Nige- ria Recorded 600 Trillion E-Payment Transactions in 2023. Technical Report, Lagos, Nige- ria: NIBSS, 2023.
[2] Nigeria Inter-Bank Settlement System. 2023 Annual Fraud Landscape Report. Technical Report, Lagos, Nigeria: NIBSS, 2024.
[3] S. Bhattacharyya, S. Jha, K. Tharakunnel, J. C. Westland. Data Mining for Credit Card Fraud: A Comparative Study. Decision Support Systems.50(3)(2021), 602–613.
[4] Randhawa, K.; Loo, C. K.; Seera, M.; Lim, C. P.; Nandi, A. K. Credit Card Fraud Detection Using AdaBoost and Majority Voting. IEEE Access, 6 (2018), 14277–14284.
[5] N. V. Chawla, K. W. Bowyer, L. O. Hall, W. P. Kegelmeyer. SMOTE: Synthetic Minority Over-Sampling Technique. Journal of Artificial Intelligence Research, 16 (2002), 321–357.
[6] J. O. Awoyemi, A. O. Adetunmbi, S. A. Oluwadare. Credit Card Fraud Detection Using Machine Learning Techniques: A Comparative Analysis. Proceedings of ICCNI, 2017, 1–9.
[7] I. A. Ayodeji. Fraud Detection and Prevention in the Nigerian Financial Indus- try. Doctoral Dissertation, Walden University, Minneapolis, MN, USA, 2024. Avail- able online: https://scholarworks.waldenu.edu/cgi/viewcontent.cgi?article= 18284&context=dissertations (accessed on 23 October 2025).
[8] Central Bank of Nigeria. Anti-Money Laundering and Combating the Financing of Ter- rorism (AML/CFT) Guidelines for Financial Institutions. Regulatory Guideline, Abuja, Nigeria: CBN, 2022.
[9] National Information Technology Development Agency. Nigeria Data Protection Regulation (NDPR) 2019. Regulatory Framework, Abuja, Nigeria: NITDA, 2019.
[10] E. A. Lopez-Rojas, S. Ellegaard, S. Axelsson. PaySim: A Financial Mobile Money Simula- tor for Fraud Detection. Proceedings of the European Modeling and Simulation Symposium (EMSS), 2016, 249–255.
[11] D. Azamuke, D. Ssebuggwawo, S. Hoppenbrouwers. Analyzing Mobile Money Transac- tion Data for Fraud Detection Using Machine Learning: A Case Study Approach. In Pro- ceedings of the 2022 International Conference on Information and Communication Tech- nologies for Development (ICT4D); ACM: New York, NY, USA, 2022; pp. 1–10. DOI: 10.1145/3531056.3542774.
[12] L. Breiman. Random Forests. Machine Learning, 45(1)(2001), 5–32.
[13] S. Xuan, G. Liu, Z. Li, L. Zheng, S. Wang, C. Jiang. Random Forest for Credit Card Fraud Detection. 2018 IEEE International Conference on Networking, Sensing and Control (ICNSC), 1–6.
[14] E. G. Dada, C. G. Nwaorgu, D. O. Oyewola, A. O. OlajuyinA Systematic Lit- erature Review of Machine Learning Techniques in Financial Fraud Prevention and Detection. International Journal of Statistical Sciences, 23(3)(2023), 315–344. DOI: 10.1504/IJSSS.2023.10057287.