Development of an Artificial Intelligence and Blockchain-Based Cyber-Security Framework for Combating Financial Fraud in Nigeria’s Digital Banking Ecosystem

Financial fraud remains a significant threat to Nigeria’s rapidly expanding digital banking ecosystem, resulting in substantial financial losses, reduced customer trust, and systemic vulnerabilities. This study developed and experimentally validated a hybrid Artificial Intelligence (AI) and Blockchain-based cybersecurity framework designed to detect, prevent, and mitigate financial fraud in real time. A dataset comprising 52,480,000 anonymized digital banking transactions, including 146,520 confirmed fraudulent cases, was used for model development and validation. The AI engine integrated Extreme Gradient Boosting, Deep Neural Networks, and Long Short-Term Memory architectures to capture structured and sequential fraud patterns, while a permissioned blockchain layer ensured transaction immutability, transparency, and tamper resistance through distributed ledger validation and smart contracts. Experimental results demonstrated detection accuracy of 96.8%, recall of 94.1%, precision of 95.2%, and a false positive rate of 3.2%, significantly outperforming existing institutional systems (p < 0.001). The model achieved an AUC score of 0.981, indicating excellent discriminatory capability. Regression analysis identified transaction velocity, device-switch frequency, geolocation deviation, and blockchain hash mismatch as significant fraud predictors. Blockchain stress testing confirmed scalability up to 3,800 transactions per second with 100% tamper detection accuracy. The findings demonstrate that integrating AI-driven behavioral analytics with blockchain-based data integrity mechanisms produces a robust, scalable, and secure cybersecurity framework capable of substantially reducing fraud risk within Nigeria’s digital financial sector. The proposed framework offers a replicable model for emerging economies seeking technologically advanced and decentralized fraud mitigation strategies.

Artificial Intelligence, Blockchain-Based, Cyber-security, Financial Fraud, Digital Banking

1. Bahnsen, A. C.s, Aouada, D., & Ottersten, B. (2016). Example-dependent cost-sensitive decision trees for credit card fraud detection. Expert Systems with Applications, 42(19), 6609–6619. [Google Scholar] [Crossref]

2. Central Bank of Nigeria. (2021). Annual report. Abuja, Nigeria: Author. [Google Scholar] [Crossref]

3. Chen, M., Mao, S., & Liu, Y. (2014). Big data: A survey. Mobile Networks and Applications, 19(2), 171–209. [Google Scholar] [Crossref]

4. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. Cambridge, MA: MIT Press. [Google Scholar] [Crossref]

5. Goodfellow, I., Shlens, J., & Szegedy, C. (2015). Explaining and harnessing adversarial examples. International Conference on Learning Representations (ICLR). [Google Scholar] [Crossref]

6. LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. [Google Scholar] [Crossref]

7. Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system. Retrieved from https://bitcoin.org/bitcoin.pdf [Google Scholar] [Crossref]

8. Nigeria Deposit Insurance Corporation. (2022). Annual report and statement of accounts. Abuja, Nigeria: Author. [Google Scholar] [Crossref]

9. Nigeria Inter-Bank Settlement System. (2023). Fraud report. Lagos, Nigeria: Author. [Google Scholar] [Crossref]

• Financial Technology (Fintech): Current Research at The Cutting Edge
• Reforming Corporate Governance in Malaysia to Address Fraudulent Financial Reporting Cases
• Stock Market Efficiency and Economic Diversification in Nigeria and South Africa
• Financial Stability and Financial Performance of Small and Medium Tiered Deposit Taking Savings and Credit Cooperatives in Kenya.
• Regulator Sandboxes for DeFi: A Comparative Analysis of Policy Effectiveness in the EU, US, and Asia Pacific