Publications

You can also find my articles on my Google Scholar profile or my Researchgate profile.

TRIAG: Tri-Reinforced Infused Generative Agents for Financial Risk Compliance

Published in Intelligent Systems with Application, 2026

Managing financial regulatory compliance (FRC) presents significant challenges for Financial Technology (FinTech) organisations. Some challenges are due to both rapidly evolving environment and limitations of existing computational support. Although FinTech organisations continuously seek contemporary methods for Financial Risk Management, common issues persist with time-consuming and labor-intensive compliance processes. Previous Artificial Intelligence-driven approaches often lack comprehensive support for the dynamic nature of regulatory requirements. To address these gaps and leverage the increasing availability of regulatory and financial data, the study utilizes a design science research paradigm to design TRIAG (Tri-Reinforced Infused Generative Agents), an innovative computational framework grounded in Multi-Agent Reinforcement Learning (MARL). The TRIAG artifact is a prototype featuring three distinct Generative AI agents that autonomously acquire, refine, and coordinate domain-specific expertise related to FinTech regulatory compliance. The study’s evaluation, involving industry professionals in a workshop setting, demonstrates that TRIAG effectively enhances the efficiency and accuracy of compliance officers decision-making processes for FRC management tasks. This work introduces a novel MARL-guided, multi-GenAI agent system applied specifically to financial regulatory risk.

Recommended citation: Sheikh, MD Rafsun and Miah, Shah J., TRIAG: Tri-Reinforced Infused Generative Agents for Financial Risk Compliance. Available at SSRN: https://ssrn.com/abstract=5798929 or http://dx.doi.org/10.2139/ssrn.5798929
Download Paper

GenRL FinTech: Supporting the Risk Management Process through Reinforcement Intelligence

Published in Discover Artificial Intelligence, 2026

Bringing technical innovations to managing financial risks has been a significant issue for managers in FinTech (financial technologies) organizations. Although FinTech organizations always explore to find new methods of Financial Risk Management (FRM), specifically for achieving smooth governance, common issues exist with time-consuming and labor-intensive processes that require adequate computational support. Previous AI (artificial intelligence) driven approaches in FRM do not fully support critical computational provisions for regulatory compliance. To address the issues, utilizing a design science research paradigm, this paper introduces a new innovative generative AI framework called ‘GenRL’ (Generative Reinforcement Learning), as an innovative computational FRM model grounded in Reinforcement Learning (RL). The GenRL artifact is a prototype featuring multiple GenAI agents that autonomously acquire and refine domain-specific expertise in FinTech regulatory compliance. Our evaluation demonstrates that GenRL enhances the efficiency of compliance officers, particularly in terms of the accuracy of FRM decision-making.

Recommended citation: Rafsun Sheikh, Shah J Miah, Peter Cook et al. GenRL FinTech: Supporting the Risk Management Process through Reinforcement Intelligence, 15 September 2025, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-7190065/v1]
Download Paper

Blockchain technology in the banking sector: a content analysis

Published in Frontiers in Blockchain, 2025

Applications of Blockchain technology (BT) offer transformative innovations in organizations. Because of its effectiveness as an intermediary-free platform, researchers consider this technological platform to adopt disruptive developments. In banking sector, BT has been adopted massively for significant disruptions, but their landscape of studies to develop general understanding are still at its emergent stage, therefore it is imperative to define existing landscape of BT for greater benefits in the research community. This paper examines existing studies of BT adoption in banking sector, with a special focus to reveal on how BT architectures can bring disruptions. Methods The analysis has scrutizised 214 relevant articles from peer-reviewed journals across four vital databases (coverage from 2021 to 15 July 2025), through an intelligent review that represents a combined iterative approach adopting both methods of Latent Dirichlet Allocation (LDA) topic modelling and content analysis. Results From an information systems viewpoint, the study divided the findings into three phases: pre-adoption, adoption, and post-adoption, highlighting blockchain’s dimensions, applications in banking, the current banking landscape, and the challenges that inhibit widespread adoption of BT in banking systems. Discussion/Conclusion The synthesized findings indicate interesting directions for future research.

Recommended citation: Sheikh, R., Miah, S. J., Skinner, J., & Cook, P. (2025). Blockchain technology in the banking sector: A content analysis. Frontiers in Blockchain, 8, 1667848. https://doi.org/10.3389/fbloc.2025.1667848
Download Paper

Five Significant Issues of Digital Payment Systems – A Content Analysis

Published in Journal of Finance and Data Science (manuscript submitted for publication), 2024

Digital payment systems have revolutionised the financial landscape, offering unprecedented convenience and efficiency. However, relative rapid adoption has highlighted critical issues that must be addressed to ensure their sustainability and broader acceptance. This study aims to explore the problems of digital payment systems and identify significant current issues. Using four prominent electronic databases, qualifying articles published in the last eight years (from Jan 2017 – June 2024) were sourced for categorisation and analysis. A total of 111 articles on issues of digital payment systems were selected for relevant content on FinTech (financial technology) studies. The research team deeply investigated the articles using established content analysis protocols. The analysis identified five vital issues in digital payment systems: security, convenience, trust, privacy, and adoption of digital payment technology. Other key issues are also identified through the content analysis, suggesting topical research priorities for bringing attention to future research.

Download Paper

An efficient transfer learning model for predicting forged (handwritten) signature

Published in 2021 International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering (IC4ME2), 2021

Signature fraud around the world is increasing at an alarming rate. Fraud in the signature may harm a person or an organization by false transactions and false document authorization, which may lead to an irreversible loss. Thus, the purpose of this research is to predict forged signature using machine learning techniques. To attain the objective, different state-of-the-art machine learning models, including Neural Network, K-Nearest Neighbors, Support Vector Machine, Decision Tree, and Random Forest Classifier, were developed to classify between fraud and real signatures. The VGG-16 pre-trained model was used to improve the Neural Networks performance. As outcome, the transfer learning based Neural Network model showed the highest accuracy-96.7%, followed by Support Vector Machine (81.7%), K-Nearest Neighbors (71.7%), Random Forest (70.0%), and Decision Tree (68.3%).

Recommended citation: M. R. Sheikh, T. H. Masud, N. I. Khan and M. N. Islam, "An Efficient Transfer Learning Model for Predicting Forged (Handwritten) Signature," 2021 International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering (IC4ME2), Rajshahi, Bangladesh, 2021, pp. 1-4, doi: 10.1109/IC4ME253898.2021.9768440.
Download Paper

MD Rafsun Sheikh