Powered by OpenAIRE graph
Found an issue? Give us feedback
addClaim

Fairness em Machine Learning: Uma análise baseada na Teoria de Resposta ao Item

Authors: Thiago Paracampo de Castro; Lucas Ferraro Cardoso; Vitor Cirilo Santos; Ronnie Oliveira Alves; Regiane Kawasaki Francês;

Fairness em Machine Learning: Uma análise baseada na Teoria de Resposta ao Item

Abstract

A popularização das ferramentas de Machine Learning tornou evidente um novo problema: modelos injustos aplicados em contextos sensíveis. Diante disso, este artigo explora a utilização da Teoria de Resposta ao Item na análise de modelos treinados em dados de contexto sensível. Como estudo de caso, o dataset Credit-G foi utilizado para analisar o impacto de uma injustiça existente em si. Os dados apresentam um viés de representação de gênero, portanto, a geração de dados sintéticos foi utilizada para mitigar essa injustiça. A avaliação pelos conceitos da TRI dos parâmetros de item para os dois grupos (enviesados e mitigados) indicam que dados enviesados podem ser percebidos pela ocorrência de instâncias com discriminação negativa. Os resultados obtidos mostram que, para o conjunto mitigado, as instâncias femininas apresentaram valores médios de −1.675 de discriminação, enquanto as masculinas apresentaram 3.2.

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Upload OA version
Are you the author of this publication? Upload your Open Access version to Zenodo!
It’s fast and easy, just two clicks!