An explainable AI decision-support-system to automate loan underwriting

Abstract Widespread adoption of automated decision making by artificial intelligence (AI) is witnessed due to specular advances in computation power and improvements in optimization algorithms especially in machine learning (ML). Complex ML models provide good prediction accuracy; however, the opacity of ML models does not provide sufficient assurance for their adoption in the automation of lending decisions. This paper presents an explainable AI decision-support-system to automate the loan underwriting process by belief-rule-base (BRB). This system can accommodate human knowledge and can also learn from historical data by supervised learning. The hierarchical structure of BRB can accommodates factual and heuristic rules. The system can explain the chain of events leading to a decision for a loan application by the importance of an activated rule and the contribution of antecedent attributes in the rule. A business case study on automation of mortgage underwriting is demonstrated to show that the BRB system can provide a good trade-off between accuracy and explainability. The textual explanation produced by the activation of rules could be used as a reason for denial of a loan. The decision-making process for an application can be comprehended by the significance of rules in providing the decision and contribution of its antecedent attributes.

[1]  V. G. Perry,et al.  Automated underwriting in mortgage lending: Good news for the underserved? , 2002 .

[2]  Fernando Gomez,et al.  Combining factual and heuristic knowledge in knowledge acquisition , 1992 .

[3]  Terry Harris,et al.  Credit scoring using the clustered support vector machine , 2015, Expert Syst. Appl..

[4]  Xuyao Zhang,et al.  Generalized fuzzy soft sets theory‐based novel hybrid ensemble credit scoring model , 2018, International Journal of Finance & Economics.

[5]  Ron Sun,et al.  Robust Reasoning: Integrating Rule-Based and Similarity-Based Reasoning , 1995, Artif. Intell..

[6]  Adnan Khashman,et al.  Neural networks for credit risk evaluation: Investigation of different neural models and learning schemes , 2010, Expert Syst. Appl..

[7]  Chrysanthos Dellarocas,et al.  Credit Scoring with Social Network Data , 2014 .

[8]  Lin Ma,et al.  Mining the customer credit using hybrid support vector machine technique , 2009, Expert Syst. Appl..

[9]  Jakub M. Tomczak,et al.  Classification Restricted Boltzmann Machine for comprehensible credit scoring model , 2015, Expert Syst. Appl..

[10]  Lyn C. Thomas,et al.  Does segmentation always improve model performance in credit scoring? , 2012, Expert Syst. Appl..

[11]  Piero P. Bonissone,et al.  Automating the Underwriting of Insurance Applications , 2005, AI Mag..

[12]  Sheng-Tun Li,et al.  The evaluation of consumer loans using support vector machines , 2006, Expert Syst. Appl..

[13]  R. Aitken.  ‘All data is credit data’: Constituting the unbanked , 2017 .

[14]  Loris Nanni,et al.  An experimental comparison of ensemble of classifiers for bankruptcy prediction and credit scoring , 2009, Expert Syst. Appl..

[15]  Wojciech Samek,et al.  Methods for interpreting and understanding deep neural networks , 2017, Digit. Signal Process..

[16]  Byeong Ho Kang,et al.  Investigation and improvement of multi-layer perception neural networks for credit scoring , 2015, Expert Syst. Appl..

[17]  Jian-Bo Yang,et al.  Belief rule-base inference methodology using the evidential reasoning Approach-RIMER , 2006, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[18]  Jian Ma,et al.  Two credit scoring models based on dual strategy ensemble trees , 2012, Knowl. Based Syst..

[19]  Wei-Wen Wu,et al.  Improving Classification Accuracy and Causal Knowledge for Better Credit Decisions , 2011, Int. J. Neural Syst..

[20]  Francisco Javier García Castellano,et al.  Expert Systems With Applications , 2022 .

[21]  C ONG,et al.  Building credit scoring models using genetic programming , 2005, Expert Syst. Appl..

[22]  Jean Dezert,et al.  UNM Digital Repository UNM Digital Repository Fusion of Sources of Evidence with Different Importances and Fusion of Sources of Evidence with Different Importances and Reliabilities Reliabilities , 2022 .

[23]  Yu Wang,et al.  Ensemble classification based on supervised clustering for credit scoring , 2016, Appl. Soft Comput..

[24]  Alexander Binder,et al.  Explaining nonlinear classification decisions with deep Taylor decomposition , 2015, Pattern Recognit..

[25]  Gavin C. Cawley,et al.  On Over-fitting in Model Selection and Subsequent Selection Bias in Performance Evaluation , 2010, J. Mach. Learn. Res..

[26]  Desheng Dash Wu,et al.  A deep learning approach for credit scoring using credit default swaps , 2017, Eng. Appl. Artif. Intell..

[27]  E A Feigenbaum,et al.  Knowledge Engineering , 1984, Annals of the New York Academy of Sciences.

[28]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[29]  Bernd Bischl,et al.  Visualizing the Feature Importance for Black Box Models , 2018, ECML/PKDD.

[30]  Jonathan Killin.  The Management and Maintenance of an Operational KADS System Development , 1992, EKAW.

[31]  Carlos Guestrin,et al.  Model-Agnostic Interpretability of Machine Learning , 2016, ArXiv.

[32]  H Zhu,et al.  A Bayesian framework for the combination of classifier outputs , 2002, J. Oper. Res. Soc..

[33]  Stephen C. H. Leung,et al.  Vertical bagging decision trees model for credit scoring , 2010, Expert Syst. Appl..

[34]  Mohamed Elhoseny,et al.  Genetic Algorithm Based Model For Optimizing Bank Lending Decisions , 2017, Expert Syst. Appl..

[35]  Bee Wah Yap,et al.  Using data mining to improve assessment of credit worthiness via credit scoring models , 2011, Expert Syst. Appl..

[36]  A. I. Marqués,et al.  Exploring the behaviour of base classifiers in credit scoring ensembles , 2012, Expert Syst. Appl..

[37]  Chee Kian Leong.  Credit Risk Scoring with Bayesian Network Models , 2015, Computational Economics.

[38]  Iwasokun Gabriel Babatunde,et al.  Fuzzy logic-driven expert system for the diagnosis of heart failure disease , 2014, Artif. Intell. Res..

[39]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[40]  Chihli Hung,et al.  A selective ensemble based on expected probabilities for bankruptcy prediction , 2009, Expert Syst. Appl..

[41]  Jian-Bo Yang,et al.  Inferential modelling and decision making with data , 2017, 2017 23rd International Conference on Automation and Computing (ICAC).

[42]  David West,et al.  Neural network credit scoring models , 2000, Comput. Oper. Res..

[43]  Jian-Bo Yang,et al.  Inference analysis and adaptive training for belief rule based systems , 2011, Expert Syst. Appl..

[44]  Ralf Stecking,et al.  Support vector machines for classifying and describing credit applicants: detecting typical and critical regions , 2005, J. Oper. Res. Soc..

[45]  Hussein A. Abdou.  Genetic programming for credit scoring: The case of Egyptian public sector banks , 2009, Expert Syst. Appl..

[46]  William P. Wagner,et al.  Knowledge acquisition for expert systems in accounting and financial problem domains , 2002, Knowl. Based Syst..

[47]  Marijana Zekic-Susac,et al.  Modelling small-business credit scoring by using logistic regression, neural networks and decision trees , 2005, Intell. Syst. Account. Finance Manag..

[48]  Amina Adadi,et al.  Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI) , 2018, IEEE Access.

[49]  So Young Sohn,et al.  Managing loan customers using misclassification patterns of credit scoring model , 2004, Expert Syst. Appl..

[50]  Zachary Chase Lipton.  The mythos of model interpretability , 2016, ACM Queue.

[51]  Srinivas Krovvidy.  Custom DU: A Web-Based Business User-Driven Automated Underwriting System , 2008, AI Mag..

[52]  Dong-Ling Xu,et al.  Evidential reasoning rule for evidence combination , 2013, Artif. Intell..

[53]  Mu-Chen Chen,et al.  Credit scoring with a data mining approach based on support vector machines , 2007, Expert Syst. Appl..

[54]  Manoj Kumar Tiwari,et al.  Computational time reduction for credit scoring: An integrated approach based on support vector machine and stratified sampling method , 2012, Expert Syst. Appl..

[55]  William P. Wagner,et al.  An experimental investigation of the impact of domain complexity on knowledge acquisition (KA) methods , 2008, Expert Syst. Appl..

[56]  Steve Cocheo.  Automated Underwriting: Tower of Techno-Babel? , 1995 .