On Discrimination Discovery Using Causal Networks

  • Lu ZhangEmail author
  • Yongkai Wu
  • Xintao Wu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9708)


Discrimination discovery is an increasingly important task in the data mining field. The purpose of discrimination discovery is to unveil discriminatory practices on the protective attribute (e.g., gender) by analyzing the dataset of historical decision records. Different types of discrimination have been proposed in the literature. We aim to develop a framework that is able to deal with all types of discrimination. We make use of the causal networks, which effectively captures the existence of discrimination patterns and can provide quantitative evidence of discrimination in decision making. In this paper, we first propose a categorization for various discrimination. Then, we present our preliminary results on four types of discrimination, namely system-level direct discrimination, the system-level indirect discrimination, group-level discrimination, and individual level discrimination. We have conducted empirical assessments on real datasets. The results show great efficacy of our approach.



This work was supported in part by U.S. National Institute of Health (1R01GM103309).


  1. 1.
    Bonchi, F., Hajian, S., Mishra, B., Ramazzotti, D.: Exposing the probabilistic causal structure of discrimination. arXiv preprint (2015). arXiv:1510.00552
  2. 2.
    Calders, T., Verwer, S.: Three naive bayes approaches for discrimination-free classification. Data Min. Knowl. Discov. 21(2), 277–292 (2010)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.: Fairness through awareness. In: Proceedings of the 3rd Innovations in Theoretical Computer Science Conference, pp. 214–226. ACM (2012)Google Scholar
  4. 4.
    Glymour, C., et al.: The TETRAD project (2004).
  5. 5.
    Hajian, S., Domingo-Ferrer, J.: A methodology for direct and indirect discrimination prevention in data mining. TKDE 25(7), 1445–1459 (2013)Google Scholar
  6. 6.
    Hajian, S., Domingo-Ferrer, J., Monreale, A., Pedreschi, D., Giannotti, F.: Discrimination-and privacy-aware patterns. Data Min. Knowl. Discov. 29(6), 1–50 (2014)MathSciNetGoogle Scholar
  7. 7.
    Kamiran, F., Calders, T.: Classifying without discriminating. In: International Conference on Computer, Control and Communication, pp. 1–6 (2009)Google Scholar
  8. 8.
    Kamiran, F., Calders, T.: Data preprocessing techniques for classification without discrimination. Knowl. Inf. Syst. 33(1), 1–33 (2012)CrossRefGoogle Scholar
  9. 9.
    Kamishima, T., Akaho, S., Sakuma, J.: Fairness-aware learning through regularization approach. In: ICDMW, pp. 643–650. IEEE (2011)Google Scholar
  10. 10.
    Lichman, M.: UCI machine learning repository (2013).
  11. 11.
    Luong, B.T., Ruggieri, S., Turini, F.: K-nn as an implementation of situation testing for discrimination discovery and prevention. In: KDD, pp. 502–510. ACM (2011)Google Scholar
  12. 12.
    Mancuhan, K., Clifton, C.: Combating discrimination using bayesian networks. Artif. Intell. Law 22(2), 211–238 (2014)CrossRefGoogle Scholar
  13. 13.
    Statistics Netherlands. Volkstelling (2001).
  14. 14.
    Pearl, J.: Comment: understanding simpsons paradox. Am. Stat. 68(1), 8–13 (2014)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Pearl, J., et al.: Causal inference in statistics: an overview. Stat. Surv. 3, 96–146 (2009)MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Pedreschi, D., Ruggieri, S., Turini, F.: Measuring discrimination in socially-sensitive decision records. In: SIAM SDM (2009)Google Scholar
  17. 17.
    Pedreshi, D., Ruggieri, S., Turini, F.: Discrimination-aware data mining. In: KDD, pp. 560–568. ACM (2008)Google Scholar
  18. 18.
    Romei, A., Ruggieri, S.: A multidisciplinary survey on discrimination analysis. Knowl. Eng. Rev. 29(05), 582–638 (2014)CrossRefGoogle Scholar
  19. 19.
    Ruggieri, S., Hajian, S., Kamiran, F., Zhang, X.: Anti-discrimination analysis using privacy attack strategies. In: Calders, T., Esposito, F., Hüllermeier, E., Meo, R. (eds.) ECML PKDD 2014, Part II. LNCS, vol. 8725, pp. 694–710. Springer, Heidelberg (2014)Google Scholar
  20. 20.
    Ruggieri, S., Pedreschi, D., Turini, F.: Data mining for discrimination discovery. ACM TKDD 4(2), 9 (2010)Google Scholar
  21. 21.
    Spirtes, P., Glymour, C.N., Scheines, R.: Causation, Prediction, and Search, vol. 81. MIT press, Cambridge (2000)zbMATHGoogle Scholar
  22. 22.
    Wu, Y., Wu, X.: Using loglinear model for discrimination discovery and prevention. Technical report, DPL-2015-002, University of Arkansas (2015)Google Scholar
  23. 23.
    Zhang, L., Wu, Y., Wu, X.: Causal Bayeisan network-based discrimination discovery and removal. Technical report, DPL-2016-001, University of Arkansas (2016)Google Scholar
  24. 24.
    Zhang, L., Yongkai, W., Xintao, W., Discovery, situation testing-based discrimination : a causal inference approach. Technical report, DPL-2016-002, University of Arkansas (2016)Google Scholar
  25. 25.
    Zliobaite, I., Kamiran, F., Calders, T.: Handling conditional discrimination. In: ICDM 2011, pp. 992–1001. IEEE (2011)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.University of ArkansasFayettevilleUSA

Personalised recommendations