In the domain of credit risk assessment lenders may have limited or no data on the historical lending outcomes of credit applicants. Typically this disproportionately affects Micro, Small, and Medium Enterprises (MSMEs), for which credit may be restricted or too costly, due to the difficulty of predicting the Probability of Default (PD). However, if data from other related credit risk domains is available Transfer Learning may be applied to successfully train models, e.g., from the credit card lending and debt consolidation (CD) domains to predict in the small business lending domain. In this article, we report successful results from an approach using transfer learning to predict the probability of default based on the novel concept of Progressive Shift Contribution (PSC) from source to target domain. Toward real-world application by lenders of this approach, we further address two key questions. The first is to explain transfer learning models, and the second is to adjust features when the source and target domains differ. To address the first question, we apply Shapley values to investigate how and why transfer learning improves model accuracy, and also propose and test a domain adaptation approach to address the second. These results show that adaptation improves model accuracy in addition to the improvement from transfer learning. We extend this by proposing and testing a combined strategy of feature selection and adaptation to convert values of source domain features to better approximate values of target domain features. Our approach includes a strategy to choose features for adaptation and an algorithm to adapt the values of these features. In this setting, transfer learning appears to improve model accuracy by increasing the contribution of less predictive features. Although the percentage improvements are small, such improvements in real world lending could be of significant economic importance.
Making responsible lending decisions involves many factors. There is a growing amount of research on machine learning applied to credit risk evaluation. This promises to enhance diversity in lending without impacting the quality of the credit available by using data on previous lending decisions and their outcomes. However, often the most accurate machine learning methods predict in ways that are not transparent to human domain experts. A consequence is increasing regulation in jurisdictions across the world requiring automated decisions to be explainable. Before the emergence of data-driven technologies lending decisions were based on human expertise, so explainable lending decisions can, in principle, be assessed by human domain experts to ensure they are fair and ethical. In this study we hypothesised that human expertise may be used to overcome the limitations of inadequate data. Using benchmark data, we investigated using machine learning on a small training set and then correcting errors in the training data with human expertise applied through Ripple-Down Rules. We found that the resulting combined model not only performed equivalently to a model learned from a large set of training data, but that the human expert’s rules also improved the decision making of the latter model. The approach is general, and can be used not only to improve the appropriateness of lending decisions, but also potentially to improve responsible decision making in any domain where machine learning training data is limited in quantity or quality.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.