Machine Learning and Credit Rating in Financial Institution in Tanzania: A Literature Review Approach
Authors:
Neema Aspedito Mfugale , Christopher Machibula
DOI: https://doi.org/10.37082/IJIRMPS.v13.i4.232545
Short DOI: https://doi.org/g9r6nq
Country: Tanzania
Full-text Research PDF File:
View |
Download
Abstract: The aim of this study was to evaluate the effectiveness of Machine Learning models in credit rating within Tanzanian financial institutions, where data scarcity and informal financial systems limit the success of traditional models. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses methodology, a systematic review of 212 studies led to the inclusion of 36 high-quality paper in the qualitative synthesis, and 16 in the literature review for analysis. The findings reveal that ML models, especially Random Forest and Gradient Boosting, outperform traditional methods in predictive accuracy and adaptability, particularly in low-data environments. These models utilize alternative data such as mobile money transactions and utility payments, making them more inclusive for underserved populations. The study concludes that Machine Learning provides a viable solution to Tanzania’s credit rating challenges and recommends adopting hybrid models and supportive regulatory frameworks to enhance credit access and financial inclusion
Keywords: Machine Learning, Credit Rating and Financial Institutions
Paper Id: 232545
Published On: 2025-07-03
Published In: Volume 13, Issue 4, July-August 2025