Datuetatik eta haietatik sortutako sailkatzaileetatik genero desbiderapenak ezabatzeko sistema. Systems to decrease gender bias in classifiers

Oier Irazabal 2020

Finished

Research line:
Machine learning
Description:

Systems built using machine learning techniques replicate the patterns that appear in the data used to learn. Data, as the society generating it, often has gender bias and, consequently, systems built applying machine learning techniques to these data will also be biased systems and will contribute to increase the bias already existing in society. This is something we should try to eliminate or even use in the opposite sense and this is the aim of the project.

Director(s):
Olatz Arbelaitz
University:
Euskal Herriko Unibertsitatea (UPV/EHU)
Center:
Informatika Fakultatea - Facultad de Informática
Department:
Konputagailuen Arkitektura eta Teknologia - Arquitectura y Tecnología de computadores
Reading year:
2020