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

Abstract

It is said that with great power comes great responsibility. Nowadays, we rely on machine learning systems that are capable of understanding text at a human-like level. Yet, relations like "man is to computer scientist what woman is to homemaker" are present in these systems.

The importance of the topic and the effect it has in the society has made it become an important research topic during the last years giving rise to different solutions.

In this work, we describe some state-of-the-art techniques that reduce gender bias in machine learning algorithms as well as assess their results employing fairness evaluation metrics.