Explaining Deep Learning Models for Credit Scoring with SHAP: A Case Study Using Open Banking Data
Journal of Risk and Financial Management · 02.04.2023 · 16 (4) · 221
DOI: 10.3390/jrfm16040221 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Risk and Financial Management, online erschienen am 02.04.2023. Verfasst von Lars Ole Hjelkrem und 1 weiteren Autor:innen. Die Arbeit wurde bisher 46-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Research Council of Norway.
Abstract (Originalsprache)
Predicting creditworthiness is an important task in the banking industry, as it allows banks to make informed lending decisions and manage risk. In this paper, we investigate the performance of two different deep learning credit scoring models developed on the textual descriptions of customer transactions available from open banking APIs. The first model is a deep learning model trained from scratch, while the second model uses transfer learning with a multilingual BERT model. We evaluate the predictive performance of these models using the area under the receiver operating characteristic curve (AUC) and Brier score. We find that a deep learning model trained from scratch outperforms a BERT transformer model finetuned on the same data. Furthermore, we find that SHAP can be used to explain such models both on a global level and for explaining rejections of actual applications.
Förderung
- The Research Council of Norway
Zitieren
APA
Hjelkrem, L. O., & Lange, P. E. D. (2023). Explaining Deep Learning Models for Credit Scoring with SHAP: A Case Study Using Open Banking Data. Journal of Risk and Financial Management, 16(4), 221. https://doi.org/10.3390/jrfm16040221
Vancouver
Hjelkrem LO, Lange PED. Explaining Deep Learning Models for Credit Scoring with SHAP: A Case Study Using Open Banking Data. Journal of Risk and Financial Management. 2023;16(4):221. doi:10.3390/jrfm16040221
BibTeX
@article{Hjelkrem2023,
author = {Hjelkrem, Lars Ole and Lange, Petter Eilif de},
title = {Explaining Deep Learning Models for Credit Scoring with SHAP: A Case Study Using Open Banking Data},
journal = {Journal of Risk and Financial Management},
year = {2023},
volume = {16},
number = {4},
pages = {221},
doi = {10.3390/jrfm16040221}
}
EUVIMED