Explainable AI for Credit Assessment in Banks

Abstract

Banks’ credit scoring models are required by financial authorities to be explainable. This paper proposes an explainable artificial intelligence (XAI) model for predicting credit default on a unique dataset of unsecured consumer loans provided by a Norwegian bank. We combined a LightGBM model with SHAP, which enables the interpretation of explanatory variables affecting the predictions. The LightGBM model clearly outperforms the bank’s actual credit scoring model (Logistic Regression). We found that the most important explanatory variables for predicting default in the LightGBM model are the volatility of utilized credit balance, remaining credit in percentage of total credit and the duration of the customer relationship. Our main contribution is the implementation of XAI methods in banking, exploring how these methods can be applied to improve the interpretability and reliability of state-of-the-art AI models. We also suggest a method for analyzing the potential economic value of an improved credit scoring model.

AI-Driven Financial Analysis: Exploring ChatGPT's Capabilities and Challenges

“…Since the 1990s, AI methodologies such as artificial neural networks, support vector machines, ensemble methods, generalized boosting, AdaBoost, and Random Forests have been employed to predict financial distress and failures in banks (Liu, Liu, & Sathye, 2021). Additionally, the application of Explainable AI (XAI) in credit models within the banking sector, such as credit scoring and credit default prediction, has been explored, contributing to the adoption of XAI techniques in the finance industry (Demajo, Vella, and Dingli, 2020;de Lange, Melsom, Vennerød, and Westgaard, 2022. The use of AI to model behavioral biases has also gained prominence. The integration of Natural Language Processing (NLP) has become increasingly vital in finance studies since the early 21st century, covering areas such as text classification, sentiment analysis, and natural language generation.…”

A framework for generating explanations of machine learning models in Fintech industry

“…AI-powered models, when paired with XAI, can provide insights into the patterns and features that led to the identification of a particular transaction as fraudulent. This knowledge empowers Fintech companies to adapt and stay ahead of new threats (De Lange et al, 2022).…”

Exploring Local Explanation of Practical Industrial AI Applications: A Systematic Literature Review

“…In the finance sector, the majority of researchers have focused on using LIME and SHAP methods with an agnostic and post hoc approach for tabular and time-series datasets (as shown in Table 2). Some researchers, including Gramegna et al [69], Benhamou et al [70], Babaei et al [72], and de Lange et al [73], have applied the SHAP method on tabular datasets, while Kumar et al [75] and Bussmann et al [74] have developed their SHAP methods on time-series datasets. Additionally, Gite et al [71] have utilized the LIME method for a tabular dataset.…”

“…Babaei et al [72] developed a Shapley-based model to predict the expected return of small and medium-sized enterprises based on their credit risk and expected profitability. Similarly, de Lange et al [73] combined SHAP with a LightGBM (light gradient-boosting machine) model to interpret explanatory variables affecting credit scoring, which outperformed the bank's logistic regression model. Additionally, Gite et al [71] proposed a model that uses long short-term memory (LSTM) and efficient machine learning techniques to accurately predict stock prices based on user sentiments derived from news headlines.…”

“…Some of the popular local explanation techniques used in industrial applications include LIME [138], SHAP [139], and Anchors [140]. These techniques have been applied to a wide range of applications such as credit risk assessment [73], image classification [141], and text classification [142]. These techniques provide transparency, increase trustworthiness, and can improve model performance.…”