Discrimination-aware data mining

ABSTRACT

In the context of civil rights law, discrimination refers to unfair or unequal treatment of people based on membership to a category or a minority, without regard to individual merit. Rules extracted from databases by data mining techniques, such as classification or association rules, when used for decision tasks such as benefit or credit approval, can be discriminatory in the above sense. In this paper, the notion of discriminatory classification rules is introduced and studied. Providing a guarantee of non-discrimination is shown to be a non trivial task. A naive approach, like taking away all discriminatory attributes, is shown to be not enough when other background knowledge is available. Our approach leads to a precise formulation of the redlining problem along with a formal result relating discriminatory rules with apparently safe ones by means of background knowledge. An empirical assessment of the results on the German credit dataset is also provided.

REFERENCES

• 1: Rakesh Agrawal , Ramakrishnan Srikant, Fast Algorithms for Mining Association Rules in Large Databases, Proceedings of the 20th International Conference on Very Large Data Bases, p.487-499, September 12-15, 1994
• 2: Rakesh Agrawal , Ramakrishnan Srikant, Privacy-preserving data mining, Proceedings of the 2000 ACM SIGMOD international conference on Management of data, p.439-450, May 15-18, 2000, Dallas, Texas, United States
• 3: G. S. Becker. The Economics of Discrimination. University of Chicago Press, 1957.
• 4: C. Clifton. Privacy preserving data mining: How do we mine data when we aren't allowed to see it? Tutorial at KDD 2003. http://www.cs.purdue.edu/homes/clifton.
• 5: B. Goethals. Frequent Itemset Mining Implementations Repository, http://fimi.cs.helsinki.fi.
• 6: H. Holzer, S. Raphael, and M. Stoll. Black job applicants and the hiring officer's race. Industrial and Labor Relations Review, 57(2):267--287, 2004.
• 7: D.H. Kaye and M. Aickin, editors. Statistical Methods in Discrimination Litigation. Marcel Dekker, Inc., 1992.
• 8: P. Kuhn. Sex discrimination in labor markets: The role of statistical evidence. The American Economic Review, 77:567--583, 1987.
• 9: M. LaCour-Little. Discrimination in mortgage lending: A critical review of the literature. J. of Real Estate Literature, 7:15--50, 1999.
• 10: B. Liu, W. Hsu, and Y. Ma. Integrating classification and association rule mining. In Proc. of KDD 1998, pages 80--86. AAAI Press, 1998.
• 11: D.J. Newman, S. Hettich, C.L. Blake, and C.J. Merz. UCI repository of machine learning databases, 1998. http://archive.ics.uci.edu/ml.
• 12: D. Pedreschi, S. Ruggieri, and F. Turini. Discrimination-aware data mining. Tech. Rep. 07-19, Dip. Inf., Univ. of Pisa, 2007. http://compass2.di.unipi.it/TR.
• 13: J. Rauch and M. Simunek. Mining for association rules by 4ft-Miner. In Proc. of INAP 2001, pages 285--295. Prolog Association of Japan, 2001. http://lispminer.vse.cz.
• 14: G. D. Squires. Racial profiling, insurance style: Insurance redlining and the uneven development of metropolitan areas. J. of Urban Affairs, 25(4):391--410, 2003.
• 15: Latanya Arvette Sweeney , Hal Abelson, Computational disclosure control: a primer on data privacy protection, Massachusetts Institute of Technology, 2001
• 16: Pang-Ning Tan , Vipin Kumar , Jaideep Srivastava, Selecting the right objective measure for association analysis, Information Systems, v.29 n.4, p.293-313, June 2004 [doi>10.1016/S0306-4379(03)00072-3]
• 17: U.S. Federal Legislation. http://www.usdoj.gov.
• 18: X. Yin and J. Han. CPAR: Classification based on Predictive Association Rules. In Proc. of SIAM DM 2003, SIAM, 2003.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications
          Subjects: Data mining

General Terms:
Algorithms, Economics, Legal Aspects

Keywords:
classification rules, discrimination