International journal of advanced smart convergence

The Institute of Internet, Broadcasting and Communication (한국인터넷방송통신학회)
권호 표지
DOI QR코드

DOI QR Code

Experimental Analysis of Bankruptcy Prediction with SHAP framework on Polish Companies

  • Tuguldur, Enkhtuya (Department of Computer Engineering, Dongseo University) ;
  • Dae-Ki, Kang (Department of Computer Engineering, Dongseo University)
  • Received : 2022.12.18
  • Accepted : 2022.12.25
  • Published : 2023.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

With the fast development of artificial intelligence day by day, users are demanding explanations about the results of algorithms and want to know what parameters influence the results. In this paper, we propose a model for bankruptcy prediction with interpretability using the SHAP framework. SHAP (SHAPley Additive exPlanations) is framework that gives a visualized result that can be used for explanation and interpretation of machine learning models. As a result, we can describe which features are important for the result of our deep learning model. SHAP framework Force plot result gives us top features which are mainly reflecting overall model score. Even though Fully Connected Neural Networks are a "black box" model, Shapley values help us to alleviate the "black box" problem. FCNNs perform well with complex dataset with more than 60 financial ratios. Combined with SHAP framework, we create an effective model with understandable interpretation. Bankruptcy is a rare event, then we avoid imbalanced dataset problem with the help of SMOTE. SMOTE is one of the oversampling technique that resulting synthetic samples are generated for the minority class. It uses K-nearest neighbors algorithm for line connecting method in order to producing examples. We expect our model results assist financial analysts who are interested in forecasting bankruptcy prediction of companies in detail.

Keywords

Acknowledgement

This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 1711122927) and the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2022R1A2C2012243).

References

  1. S.R. Islam, W. Eberle, S. Bundy and S.K. Ghafoor, "Infusing Domain Knowledge in AI-based "Black Box" Models for Better Explainability with Application in Bankruptcy Prediction," in Proc. ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2019, Anomaly Detection in Finance Workshop, Anchorage AK USA Aug. 4-8, 2019. DOI: https://doi.org/10.48550/arXiv.1905.11474
  2. C. Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (2nd ed.), Independently published, 2022.
  3. A. Roth, The Shapley Value, Cambridge University Press, 1988.
  4. D. Unzueta, "Fully Connected Layer vs. Convolutional Layer: Explained." Oct. 18, 2022 . builtin.com/machinelearning/fully-connected-layer, Accessed Feb. 17, 2023.
  5. O. Biran and C. Cotton, "Explanation and Justification in Machine Learning: A Survey," in Proc. IJCAI Workshop on Explainable Artificial Intelligence (XAI), Aug. 13, 2017.
  6. J.L. Bellovary, S. Giacomino, and M.D. Akers, "A Review of Bankruptcy Prediction Studies: 1930-Present," Journal of Financial Education, Vol. 33, pp. 1-42, January 2007.
  7. E. Sfakianakis, "Bankruptcy Prediction Model for Listed Companies in Greece," Investment Management and Financial Innovations, Vol. 18, No. 2, pp. 166-180, May 2021. DOI: http://dx.doi.org/10.21511/imfi.18(2).2021.14
  8. S. Tian, Y. Yu, and H. Guo, "Variable Selection and Corporate Bankruptcy Forecasts," Journal of Banking & Finance, Vol. 52, pp. 89-100, March 2015. DOI: https://doi.org/10.1016/j.jbankfin.2014.12.003
  9. L. Cultrera, and X. Bredart, "Bankruptcy Prediction: The Case of Belgian SMEs," Review of Accounting and Finance, Vol. 15, No. 1, pp. 101-119. February 2016. DOI: https://doi.org/10.1108/RAF-06-2014-0059
  10. B. Ramsundar, and R.B. Zadeh, "TensorFlow for Deep Learning," Chapter 4. Fully Connected Deep Networks, March 2018.
  11. L.F.S. Scabini, and O.M. Bruno, "Structure and Performance of Fully Connected Neural Networks: Emerging Complex Network Properties," arXiv:2107.14062v1, July 2021. DOI: https://doi.org/10.48550/arXiv.2107.14062
  12. B. Rozemberczki, L.Watson, P.Bayer, H.T.Yang, Oliver Kiss, S. Nilsson and R. Sarkar "The Shapley Value in Machine Learning," in Proc. 31st International Joint Conference on Artificial Intelligence (IJCAI-22), International Joint Conferences on Artificial Intelligence Organization, pp. 5572-5579. Feb 11, 2022. DOI: https://doi.org/10.48550/arXiv.2202.05594
  13. S.M. Lundberg, and S.-I. Lee, "A Unified Approach to Interpreting Model Predictions," in Proc. 31st International Conference on Neural Information Processing Systems, pp. 4768-4777, December 2017. DOI: https://doi.org/10.48550/arXiv.1705.07874
  14. N.V. Chawla, K.W. Bowyer, L.O. Hall, and W. P. Kegelmeyer, "SMOTE: Synthetic Minority Over-sampling Technique," Vol. 16, No. 1, pp. 321-357, June 2002. DOI: https://doi.org/10.1613/jair.953