Published in West Africa Fintech & Digital Economy Conference 2025. This article was peer reviewed before publication.
Gulf of Guinea Business School
Explainable credit scoring from alternative data for thin-file borrowers
Abstract
Thin-file borrowers lack the credit history that conventional scoring requires. We build a credit-scoring model from airtime top-ups, mobile-money flows, and handset metadata, pairing gradient-boosted trees with SHAP-based explanations so that adverse-action reasons can be disclosed. On a portfolio of 60,000 microloans the model improved default discrimination while producing human-readable reasons for each decision. We discuss fairness auditing across gender and region.
Keywords: Credit scoring, Alternative data, Explainable AI, Microfinance
Cite as: Kwame Mensah & Ama Owusu (2025). Explainable credit scoring from alternative data for thin-file borrowers. West Africa Fintech & Digital Economy Conference 2025. https://quorumpress.org/papers/explainable-credit-scoring-from-83aw4
© 2025 Mensah and Owusu. Open access under CC BY 4.0.
An abstract submission. There is no manuscript body.
Cite this paper
Kwame Mensah, Ama Owusu. Explainable credit scoring from alternative data for thin-file borrowers. West Africa Fintech & Digital Economy Conference 2025, 2025. https://quorumpress.org/papers/explainable-credit-scoring-from-83aw4
Publication ID: 83aw4