Research Article

Integrating Blockchain and Data Analytics to Strengthen Financial Traceability and Anti-Fraud Controls

Authors

  • Md Shoriful Islam Chowdhury Department of Public Administration, University of Chittagong, Chattogram, Bangladesh
  • Kaniz Sultana Chy Department of Information Systems, Lamar University, Beaumont, Texas, USA
  • Md Ashiqul Islam College of Graduate and Professional Studies, Trine University, Angola, Indiana, USA
  • Md Nurul Islam Chowdhury Senior Principal Officer, Social Islami Bank PLC, Chattogram, Bangladesh; Department of Economics, University of Chittagong, Chattogram, Bangladesh

Abstract

This study presents and assesses an Integrated Blockchain Analytics Framework that uses blockchain data to identify illegal activity related to ransomware attacks by combining on-chain transactional data with off-chain regulatory and institutional knowledge. Blockchain-based ledgers provide a transparent view of each individual transaction; however, due to the pseudonymous nature of most users of cryptocurrency, along with the fact that much of this information may be fragmented between the blockchain and other institutions' databases, makes it difficult to attribute and enforce Anti-Money Laundering (AML) requirements. In this research, we propose an integrated framework for analyzing blockchain data using a combination of graph-based modeling of transactions, data enrichment and explainable machine learning techniques to enhance the traceability and compliance analysis of financial activity. The proposed framework includes a structured pipeline for preprocessing and feature engineering of the data, as well as, an interpretable risk score for the purpose of supporting both the regulatory review process and the workflow of investigators. The results also demonstrate the need to combine explainable analytics with blockchain forensic techniques to increase transparency, reproducibility and usability by regulators. Utilizing a publicly available labeled dataset of Bitcoin transactions from ransomware attacks, the approach shows significant increases in the completeness of the traceability, reductions in the time required to detect suspicious activity and efficiencies in the analysis of large volumes of data when comparing our approach to traditional rule-based approaches used for monitoring. Overall, the results indicate that a hybrid, explainable, blockchain analytic technique could significantly improve the effectiveness of AML and help meet U.S. government policy objectives concerning the risks associated with the use of digital assets and the integrity of the U.S. financial system.

Article information

Journal

Journal of Computer Science and Technology Studies

Volume (Issue)

5 (3)

Pages

189-202

Published

2023-08-09

How to Cite

Md Shoriful Islam Chowdhury, Kaniz Sultana Chy, Md Ashiqul Islam, & Md Nurul Islam Chowdhury. (2023). Integrating Blockchain and Data Analytics to Strengthen Financial Traceability and Anti-Fraud Controls. Journal of Computer Science and Technology Studies, 5(3), 189-202. https://doi.org/10.32996/jcsts.2023.5.3.14

Downloads

Views

0

Downloads

0

Keywords:

Blockchain Analytics, Financial Traceability, Ransomware Detection, Anti-Money Laundering, Machine Learning Risk Scoring, U.S. Financial Compliance