Ethical and socially-aware data labels

E. Beretta, A. Vetrò, B. Lepri, J.C. De Martin

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

© 2019, Springer Nature Switzerland AG.Many software systems today make use of large amount of personal data to make recommendations or decisions that affect our daily lives. These software systems generally operate without guarantees of non-discriminatory practices, as instead often required to human decision-makers, and therefore are attracting increasing scrutiny. Our research is focused on the specific problem of biased software-based decisions caused from biased input data. In this regard, we propose a data labeling framework based on the identification of measurable data characteristics that could lead to downstream discriminating effects. We test the proposed framework on a real dataset, which allowed us to detect risks of discrimination for the case of population groups.
Original languageEnglish
Title of host publicationInformation Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings
EditorsJ.A. Lossio-Ventura, D. Muñante, H. Alatrista-Salas
PublisherSpringer Verlag
Pages320-327
ISBN (Print)9783030116798
DOIs
Publication statusPublished - 2019
Externally publishedYes
Event5th International Conference on Information Management and Big Data, SIMBig 2018 - Lima, Peru
Duration: 3 Sept 20185 Sept 2018

Publication series

NameCommunications in Computer and Information Science
ISSN (Print)1865-0929

Conference

Conference5th International Conference on Information Management and Big Data, SIMBig 2018
Country/TerritoryPeru
CityLima
Period3/09/185/09/18

Fingerprint

Dive into the research topics of 'Ethical and socially-aware data labels'. Together they form a unique fingerprint.

Cite this