Abstract :

Money Laundering Poses A Severe Threat To Financial Integrity, Enabling Criminal Enterprises To Disguise Illicit Funds And Evade Detection. This Project, Tracing Money Laundering With Statistics And Machine Learning, Develops A Hybrid Framework That Combines Statistical Analytics, Graphbased Techniques, And Supervised/unsupervised Machine Learning To Identify, Trace, And Prioritize Suspicious Financial Flows. The System First Applies Statistical Anomaly Detection And Time-series Analysis To Flag Unusual Transaction Patterns (volume Spikes, Velocity Changes, Structuring). Suspicious Accounts And Transactions Are Then Modeled As Nodes And Edges In Transaction Graphs; Graph Features (centrality, Community Membership, Cyclicity) And Engineered Transactional Features Feed Into Machine Learning Models—classification (e.g., Random Forest, XGBoost) For Known-risk Prediction And Clustering/anomaly Detection (e.g., DBSCAN, Autoencoders) For Discovering Novel Laundering Schemes. Explainable AI Methods And Rule-based Overlays Ensure Transparency And Regulatory Interpretability, While Case-ranking And Visualization Tools Support Investigator Workflows. Evaluated On Synthetic And Real-world Datasets, The Hybrid Approach Improves Detection Rates And Reduces False Positives Compared To Baseline Rule-only Systems. The Framework Is Designed For Scalable Deployment In Banks And Regulators To Strengthen AML (Anti-Money Laundering) Efforts And Accelerate Forensic Investigations. Keywords: Anti-Money Laundering, Anomaly Detection, Transaction Graphs, Machine Learning, Statistical Methods, Explainable AI, Financial Forensics.

Published:

28-10-2025

Issue:

Vol. 25 No. 10 (2025)

Page Nos:

222-227

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite