# Three naive Bayes approaches for discrimination-free classification

@article{Calders2010ThreeNB, title={Three naive Bayes approaches for discrimination-free classification}, author={Toon Calders and Sicco Verwer}, journal={Data Mining and Knowledge Discovery}, year={2010}, volume={21}, pages={277-292} }

In this paper, we investigate how to modify the naive Bayes classifier in order to perform classification that is restricted to be independent with respect to a given sensitive attribute. Such independency restrictions occur naturally when the decision process leading to the labels in the data-set was biased; e.g., due to gender or racial discrimination. This setting is motivated by many cases in which there exist laws that disallow a decision that is partly based on discrimination. Naive…

## Topics from this paper

## 594 Citations

Introducing Positive Discrimination in Predictive Models

- History, Computer ScienceDiscrimination and Privacy in the Information Society
- 2013

In this chapter, three solutions for the discrimination-aware classification problem that are based upon Bayesian classifiers are given that can be seen as ways to introduce positive discrimination.

Techniques for Discrimination-Free Predictive Models

- Computer Science, EngineeringDiscrimination and Privacy in the Information Society
- 2013

This chapter studies a case of income prediction, where the available historical data exhibits a wage gap between the genders, and the discrimination-aware techniques succeed in significantly reducing gender discrimination without impairing too much the accuracy.

Handling Conditional Discrimination

- Computer Science, Engineering2011 IEEE 11th International Conference on Data Mining
- 2011

This work develops local techniques for handling conditional discrimination when one of the attributes is considered to be explanatory, and demonstrates that the new local techniques remove exactly the bad discrimination, allowing differences in decisions as long as they are explainable.

Discrimination Aware Decision Tree Learning

- Computer Science2010 IEEE International Conference on Data Mining
- 2010

Experimental evaluation shows that the proposed approach advances the state-of-the-art in the sense that the learned decision trees have a lower discrimination than models provided by previous methods, with little loss in accuracy.

Decision Theory for Discrimination-Aware Classification

- Computer Science2012 IEEE 12th International Conference on Data Mining
- 2012

The first and second solutions exploit the reject option of probabilistic classifier(s) and the disagreement region of general classifier ensembles to reduce discrimination and relate both solutions with decision theory for better understanding of the process.

Data preprocessing techniques for classification without discrimination

- Computer ScienceKnowledge and Information Systems
- 2011

This paper surveys and extends existing data preprocessing techniques, being suppression of the sensitive attribute, massaging the dataset by changing class labels, and reweighing or resampling the data to remove discrimination without relabeling instances and presents the results of experiments on real-life data.

Learning Fair Naive Bayes Classifiers by Discovering and Eliminating Discrimination Patterns

- Computer Science, MathematicsAAAI
- 2020

This paper proposes an algorithm to discover and mine for discrimination patterns in a naive Bayes classifier, and shows how to learn maximum likelihood parameters subject to these fairness constraints.

Exploiting reject option in classification for social discrimination control

- Computer ScienceInf. Sci.
- 2018

This paper utilizes reject option in classification, a general decision theoretic framework for handling instances whose labels are uncertain, for modeling and controlling discriminatory decisions, and presents three different solutions for discrimination-aware classification that are effective at removing illegal discrimination from the predictions.

Quantifying explainable discrimination and removing illegal discrimination in automated decision making

- Computer ScienceKnowledge and Information Systems
- 2012

The refined notion of conditional non-discrimination in classifier design is introduced and it is shown that some of the differences in decisions across the sensitive groups can be explainable and are hence tolerable.

Combating discrimination using Bayesian networks

- Computer ScienceArtificial Intelligence and Law
- 2014

A discrimination discovery method based on modeling the probability distribution of a class using Bayesian networks and a classification method that corrects for the discovered discrimination without using protected attributes in the decision process are proposed.

## References

SHOWING 1-10 OF 15 REFERENCES

Classifying without discriminating

- Computer Science2009 2nd International Conference on Computer, Control and Communication
- 2009

This paper proposes a new classification scheme for learning unbiased models on biased training data based on massaging the dataset by making the least intrusive modifications which lead to an unbiased dataset and learns a non-discriminating classifier.

Discrimination-aware data mining

- Computer ScienceKDD
- 2008

This 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, and an empirical assessment of the results on the German credit dataset.

The Foundations of Cost-Sensitive Learning

- Computer ScienceIJCAI
- 2001

It is argued that changing the balance of negative and positive training examples has little effect on the classifiers produced by standard Bayesian and decision tree learning methods, and the recommended way of applying one of these methods is to learn a classifier from the training set and then to compute optimal decisions explicitly using the probability estimates given by the classifier.

Data mining for discrimination discovery

- Computer ScienceTKDD
- 2010

This article formalizes the processes of direct and indirect discrimination discovery by modelling protected-by-law groups and contexts where discrimination occurs in a classification rule based syntax and proposes two inference models and provides automatic procedures for their implementation.

Classification with No Discrimination by Preferential Sampling

- 2010

The concept of classification without discrimination is a new area of research. (Kamiran & Calders, ) introduced the idea of Classification with No Discrimination (CND)and proposed a solution based…

Measuring Discrimination in Socially-Sensitive Decision Records

- Computer ScienceSDM
- 2009

This work tackles the problem of determining, given a dataset of historical decision records, a precise measure of the degree of discrimination suffered by a given group in a given context with respect to the decision, and introduces a collection of quantitative measures of discrimination.

Learning decision trees for loss minimization in multi-class problems

- Computer Science
- 1999

This work studies methods for modifying C4.5 to incorporate arbitrary loss matrices and tests several methods: a wrapper method and some simple heuristics, and shows that this measure can predict when more e cient methods will be applied and when the wrapper method must be applied.

Nearest Neighbour Classification with Monotonicity Constraints

- Mathematics, Computer ScienceECML/PKDD
- 2008

A modified nearest neighbour algorithm for the construction of monotone classifiers from data is proposed that often outperforms standard kNN in problems where the monotonicity constraints are applicable.

Building Classifiers with Independency Constraints

- Computer Science2009 IEEE International Conference on Data Mining Workshops
- 2009

This paper studies the classification with independency constraints problem: find an accurate model for which the predictions are independent from a given binary attribute and proposes two solutions and presents an empirical validation.

Statistical Model for Rough Set Approach to Multicriteria Classification

- Mathematics, Computer SciencePKDD
- 2007

A new approach to variable consistency that is based on maximum likelihood estimation is introduced and the equivalence of the variable consistency rough sets to the specific risk-minimizing decision rule in statistical decision theory is shown.