Machine learning framework targets money laundering in Rwandan mobile money

Researchers have developed a machine learning framework to detect money laundering and terrorism financing within Rwanda's mobile money ecosystem. The framework addresses challenges such as extreme class imbalance, delayed transaction labels, and limited investigator capacity. It was evaluated using a synthetic dataset of over 9.5 million transactions, benchmarking various supervised and unsupervised models, with a fusion stacker achieving the highest performance. AI

IMPACT This framework offers a potential solution for regulators to combat financial crime in emerging digital economies.

RANK_REASON Academic paper detailing a machine learning framework for a specific real-world problem. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• Bharti Airtel
• Financial Intelligence Centre (FIC)
• LightGBM
• money-laundering
• National Bank of Rwanda
• Rwanda
• sub-Saharan Africa
• terrorism-financing

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →