fairness: Algorithmic Fairness Metrics

Offers various metrics of algorithmic fairness. Fairness in machine learning is an emerging topic with the overarching aim to critically assess algorithms (predictive and classification models) whether their results reinforce existing social biases. While unfair algorithms can propagate such biases and offer prediction or classification results with a disparate impact on various sensitive subgroups of populations (defined by sex, gender, ethnicity, religion, income, socioeconomic status, physical or mental disabilities), fair algorithms possess the underlying foundation that these groups should be treated similarly / should have similar outcomes. The fairness R package offers the calculation and comparisons of commonly and less commonly used fairness metrics in population subgroups. These methods are described by Calders and Verwer (2010) <doi:10.1007/s10618-010-0190-x>, Chouldechova (2017) <doi:10.1089/big.2016.0047>, Feldman et al. (2015) <doi:10.1145/2783258.2783311> , Friedler et al. (2018) <doi:10.1145/3287560.3287589> and Zafar et al. (2017) <doi:10.1145/3038912.3052660>. The package also offers convenient visualizations to help understand fairness metrics.

Version:	1.1.1
Depends:	R (≥ 3.5.0)
Imports:	caret, devtools, e1071, ggplot2, pROC
Suggests:	testthat, knitr, rmarkdown
Published:	2020-07-26
Author:	Nikita Kozodoi [aut, cre], Tibor V. Varga [aut]
Maintainer:	Nikita Kozodoi <n.kozodoi at icloud.com>
BugReports:	https://github.com/kozodoi/fairness/issues
License:	MIT + file LICENSE
NeedsCompilation:	no
Language:	en-US
Materials:	NEWS
CRAN checks:	fairness results

Downloads:

Linking:

Please use the canonical form https://CRAN.R-project.org/package=fairness to link to this page.