IDEAS home Printed from https://ideas.repec.org/a/eee/ejores/v317y2024i2p273-285.html
   My bibliography  Save this article

Supervised feature compression based on counterfactual analysis

Author

Listed:
  • Piccialli, Veronica
  • Romero Morales, Dolores
  • Salvatore, Cecilia

Abstract

Counterfactual Explanations are becoming a de-facto standard in post-hoc interpretable machine learning. For a given classifier and an instance classified in an undesired class, its counterfactual explanation corresponds to small perturbations of that instance that allows changing the classification outcome. This work aims to leverage Counterfactual Explanations to detect the important decision boundaries of a pre-trained black-box model. This information is used to build a supervised discretization of the features in the dataset with a tunable granularity. Using the discretized dataset, an optimal Decision Tree can be trained that resembles the black-box model, but that is more interpretable and compact. Numerical results on real-world datasets show the effectiveness of the approach in terms of accuracy and sparsity.

Suggested Citation

  • Piccialli, Veronica & Romero Morales, Dolores & Salvatore, Cecilia, 2024. "Supervised feature compression based on counterfactual analysis," European Journal of Operational Research, Elsevier, vol. 317(2), pages 273-285.
    • Handle: RePEc:eee:ejores:v:317:y:2024:i:2:p:273-285
    DOI: 10.1016/j.ejor.2023.11.019
    as

    Download full text from publisher

    File URL: http://www.sciencedirect.com/science/article/pii/S0377221723008603
    Download Restriction: Full text for ScienceDirect subscribers only
    File URL: https://libkey.io/10.1016/j.ejor.2023.11.019?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    References listed on IDEAS

    as
    1. Oktay Günlük & Jayant Kalagnanam & Minhan Li & Matt Menickelly & Katya Scheinberg, 2021. "Optimal decision trees for categorical data via integer programming," Journal of Global Optimization, Springer, vol. 81(1), pages 233-260, September.
    2. Piramuthu, Selwyn, 2004. "Evaluating feature selection methods for learning in data mining applications," European Journal of Operational Research, Elsevier, vol. 156(2), pages 483-494, July.
    3. Bastos, João A. & Matos, Sara M., 2022. "Explainable models of credit losses," European Journal of Operational Research, Elsevier, vol. 301(1), pages 386-394.
    4. Dumitrescu, Elena & Hué, Sullivan & Hurlin, Christophe & Tokpavi, Sessi, 2022. "Machine learning for credit scoring: Improving logistic regression with non-linear decision-tree effects," European Journal of Operational Research, Elsevier, vol. 297(3), pages 1178-1192.
    5. Kozodoi, Nikita & Jacob, Johannes & Lessmann, Stefan, 2022. "Fairness in credit scoring: Assessment, implementation and profit implications," European Journal of Operational Research, Elsevier, vol. 297(3), pages 1083-1094.
    6. Nikita Kozodoi & Johannes Jacob & Stefan Lessmann, 2021. "Fairness in Credit Scoring: Assessment, Implementation and Profit Implications," Papers 2103.01907, arXiv.org, revised Jun 2022.
    7. Emilio Carrizosa & Cristina Molero-Río & Dolores Romero Morales, 2021. "Mathematical optimization in classification and regression trees," TOP: An Official Journal of the Spanish Society of Statistics and Operations Research, Springer;Sociedad de Estadística e Investigación Operativa, vol. 29(1), pages 5-33, April.
    8. Fethi, Meryem Duygun & Pasiouras, Fotios, 2010. "Assessing bank efficiency and performance with operational research and artificial intelligence techniques: A survey," European Journal of Operational Research, Elsevier, vol. 204(2), pages 189-198, July.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Sullivan Hué, 2022. "GAM(L)A: An econometric model for interpretable machine learning," French Stata Users' Group Meetings 2022 19, Stata Users Group.
    2. De Bock, Koen W. & Coussement, Kristof & Caigny, Arno De & Słowiński, Roman & Baesens, Bart & Boute, Robert N. & Choi, Tsan-Ming & Delen, Dursun & Kraus, Mathias & Lessmann, Stefan & Maldonado, Sebast, 2024. "Explainable AI for Operational Research: A defining framework, methods, applications, and a research agenda," European Journal of Operational Research, Elsevier, vol. 317(2), pages 249-272.
    3. Koen W. de Bock & Kristof Coussement & Arno De Caigny & Roman Slowiński & Bart Baesens & Robert N Boute & Tsan-Ming Choi & Dursun Delen & Mathias Kraus & Stefan Lessmann & Sebastián Maldonado & David , 2023. "Explainable AI for Operational Research: A Defining Framework, Methods, Applications, and a Research Agenda," Post-Print hal-04219546, HAL.
    4. Emmanuel Flachaire & Gilles Hacheme & Sullivan Hu'e & S'ebastien Laurent, 2022. "GAM(L)A: An econometric model for interpretable Machine Learning," Papers 2203.11691, arXiv.org.
    5. Li, Zhe & Liang, Shuguang & Pan, Xianyou & Pang, Meng, 2024. "Credit risk prediction based on loan profit: Evidence from Chinese SMEs," Research in International Business and Finance, Elsevier, vol. 67(PA).
    6. Jie Shi & Arno P. J. M. Siebes & Siamak Mehrkanoon, 2023. "TransCORALNet: A Two-Stream Transformer CORAL Networks for Supply Chain Credit Assessment Cold Start," Papers 2311.18749, arXiv.org.
    7. Sun, Weixin & Zhang, Xuantao & Li, Minghao & Wang, Yong, 2023. "Interpretable high-stakes decision support system for credit default forecasting," Technological Forecasting and Social Change, Elsevier, vol. 196(C).
    8. Silvana M. Pesenti & Pietro Millossovich & Andreas Tsanakas, 2023. "Differential Sensitivity in Discontinuous Models," Papers 2310.06151, arXiv.org.
    9. Kraus, Mathias & Tschernutter, Daniel & Weinzierl, Sven & Zschech, Patrick, 2024. "Interpretable generalized additive neural networks," European Journal of Operational Research, Elsevier, vol. 317(2), pages 303-316.
    10. Topuz, Kazim & Urban, Timothy L. & Yildirim, Mehmet B., 2024. "A Markovian score model for evaluating provider performance for continuity of care—An explainable analytics approach," European Journal of Operational Research, Elsevier, vol. 317(2), pages 341-351.
    11. Anna Langenberg & Shih-Chi Ma & Tatiana Ermakova & Benjamin Fabian, 2023. "Formal Group Fairness and Accuracy in Automated Decision Making," Mathematics, MDPI, vol. 11(8), pages 1-25, April.
    12. Henry Penikas, 2023. "Unaccounted model risk for Basel IRB models deemed acceptable by conventional validation criteria," Risk Management, Palgrave Macmillan, vol. 25(4), pages 1-25, December.
    13. Lukas Jurgensmeier & Bernd Skiera, 2023. "Measuring Self-Preferencing on Digital Platforms," Papers 2303.14947, arXiv.org, revised Feb 2024.
    14. Lu, Xuefei & Calabrese, Raffaella, 2023. "The Cohort Shapley value to measure fairness in financing small and medium enterprises in the UK," Finance Research Letters, Elsevier, vol. 58(PC).
    15. Zha, Yong & Wang, Yuting & Li, Quan & Yao, Wenying, 2022. "Credit offering strategy and dynamic pricing in the presence of consumer strategic behavior," European Journal of Operational Research, Elsevier, vol. 303(2), pages 753-766.
    16. Jos'e Pombal & Andr'e F. Cruz & Jo~ao Bravo & Pedro Saleiro & M'ario A. T. Figueiredo & Pedro Bizarro, 2022. "Understanding Unfairness in Fraud Detection through Model and Data Bias Interactions," Papers 2207.06273, arXiv.org.
    17. Blanquero, Rafael & Carrizosa, Emilio & Molero-Río, Cristina & Morales, Dolores Romero, 2022. "On sparse optimal regression trees," European Journal of Operational Research, Elsevier, vol. 299(3), pages 1045-1054.
    18. Victor Blanco & Alberto Japón & Justo Puerto, 2022. "Robust optimal classification trees under noisy labels," Advances in Data Analysis and Classification, Springer;German Classification Society - Gesellschaft für Klassifikation (GfKl);Japanese Classification Society (JCS);Classification and Data Analysis Group of the Italian Statistical Society (CLADAG);International Federation of Classification Societies (IFCS), vol. 16(1), pages 155-179, March.
    19. Dimitrios Nikolaidis & Michalis Doumpos, 2022. "Credit Scoring with Drift Adaptation Using Local Regions of Competence," SN Operations Research Forum, Springer, vol. 3(4), pages 1-28, December.
    20. Dangxing Chen & Weicheng Ye & Jiahui Ye, 2022. "Interpretable Selective Learning in Credit Risk," Papers 2209.10127, arXiv.org.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:eee:ejores:v:317:y:2024:i:2:p:273-285. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Catherine Liu (email available below). General contact details of provider: http://www.elsevier.com/locate/eor .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.