Operational Machine Learning for Mobile Money Laundering Detection in Rwanda
Summary
This paper develops a transaction-monitoring framework for detecting laundering in Rwandan mobile money. It addresses severe class imbalance, delayed and limited labels, and finite investigative capacity. Using a synthetic dataset of more than nine million transactions covering 17 laundering typologies, it constructs account-level behavioral features such as transaction velocity, flow direction, counterparty diversity, and bursts of activity. It benchmarks supervised classifiers, anomaly detectors, an autoencoder, and a model that combines predictions.
Evaluation focuses on operational usefulness: precision-recall area, recall at a target precision, recall within a limited alert volume, and alert rates. On a chronological test period, a fusion model slightly exceeds LightGBM in PR-AUC and achieves roughly 91% precision, while identifying 59 cases; LightGBM identifies 64 at roughly 89% precision. The authors connect score bands to analyst review and escalation workflows and propose validating the approach on real data. Results are based on synthetic data, so performance in live monitoring remains unverified; the contribution is an evaluation and governance framework, not a novel detection algorithm.
Key ideas
- The framework is designed for extreme class imbalance, scarce labels, and limited investigator capacity.
- Account-level features summarize transaction pace, flow direction, counterparties, and burstiness.
- Several supervised and anomaly-detection methods are assessed alongside a prediction-fusion model.
- Operational metrics measure detection quality under precision and alert-volume constraints.
- Reported results use synthetic data and require validation on real transactions.
Tags
Full text
# Detecting Money Laundering in Rwandan Mobile Money: A Machine Learning Framework # Detecting Money Laundering in Rwandan Mobile Money: A Machine Learning Framework Mobile money has widened financial access across Sub-Saharan Africa and enlarged the surface for money-laundering and terrorism-financing (ML/TF) activity in ecosystems dominated by high-volume, low-value transactions. Rwanda is a case in point: several million active mobile-money users, telecom-led wallets on the MTN and Airtel networks, and a Financial Intelligence Centre (FIC) supervising transaction streams whose scale exceeds static rule-based monitoring. This paper develops and evaluates a transaction-monitoring framework aligned to the Rwandan AML/CFT regime under (i) extreme class imbalance (~0.1% prevalence), (ii) scarce and delayed labels, and (iii) bounded investigator capacity. Using SAML-D, a synthetic dataset of 9,504,852 transactions with 17 laundering typologies, we engineer account-centric behavioural features (rolling velocity, net-flow directionality, counterparty diversity, burstiness) and benchmark supervised classifiers (Logistic Regression, Random Forest, LightGBM), unsupervised anomaly detectors (Isolation Forest, Local Outlier Factor), a dense autoencoder, and a late-fusion meta-learner. Evaluation is operational: PR-AUC, recall at a calibrated ~90%-precision point, recall at top-K%, and alerts per 10,000. On the chronologically held-out test period, LightGBM attains PR-AUC = 0.0469, capturing 64 laundering cases at precision ~0.89 with 0.51 alerts per 10,000; the fusion stacker reaches PR-AUC = 0.0477 at precision ~0.91 and 0.46 alerts per 10,000, recovering 59 true positives. We map score bands to Rwanda-relevant analyst workflows and STR/SAR escalation, and outline a staged path from synthetic prototyping to real-data validation with the National Bank of Rwanda and FIC. The contribution is operational: a governance-aware pipeline and evaluation protocol calibrated to the constraints of an African mobile-money regulator, not a new algorithm.
Shown in full with attribution under the source's licence. Licence: abstract CC0
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.