Ethnic classifications in algorithmic decision-making processes

PDF
paper-35.pdf - Published
Download (187kB)

URL:	https://ceur-ws.org/Vol-3442/
Additional URL:	https://ceur-ws.org/Vol-3442/paper-35.pdf
URN:	urn:nbn:de:bsz:180-madoc-657350
Document Type:	Conference or workshop publication
Year of publication:	2023
Book title:	EWAF 2023: European Workshop on Algorithmic Fairness : proceedings of the 2nd European Workshop on Algorithmic Fairness, Winterthur, Switzerland, June 7th to 9th, 2023
The title of a journal, publication series:	CEUR Workshop Proceedings
Volume:	3442
Page range:	1-5
Conference title:	EWAF 2023
Location of the conference venue:	Winterthur, Switzerland
Date of the conference:	07.-09.06.2023
Publisher:	Alvarez, Jose M. ; Fabris, Alessandro ; Heitz, Christoh ; Hertweck, Corinna ; Loi, Michele ; Zehlike, Meike
Place of publication:	Aachen, Germany
Publishing house:	RWTH Aachen
ISSN:	1613-0073
Publication language:	English
Institution:	Außerfakultäre Einrichtungen > Mannheim Centre for European Social Research - Research Department A
Pre-existing license:	Creative Commons Attribution 4.0 International (CC BY 4.0)
Subject:	310 Statistics
Abstract:	In this study, we address the challenges and implications of ensuring fairness in algorithmic decision-making (ADM) practices related to ethnicity. We provide an overview of ethnic classification schemes in European countries and emphasize how the distinct approaches to ethnicity and race in Europe can impact fairness assessments in ADM. Using German data, we train machine learning classifiers and explore the fairness implications of different ethnic classifications in labor market- and health-related prediction tasks using common group-based fairness metrics. The findings contribute to the understanding of fairness in ADM from a European perspective.