Sexismo no Brasil: Análise de um Word Embedding por meio de testes baseados em associação implícita
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Standard
BRAZILIAN SYMPOSIUM IN INFORMATION AND HUMAN LANGUAGE TECHNOLOGY (STIL). ed. / Helena Caseli; Maria Jose Bocorny Finatto. Sociedade Brasileira de Computação (SBC), 2023. p. 53-62.
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Sexismo no Brasil
T2 - 14th Brazilian Symposium in Information and Human Language Technology - STIL 2023
AU - Quadros dos Reis, Valéria
N1 - Conference code: 14
PY - 2023/9/25
Y1 - 2023/9/25
N2 - This work reports experiments based on the Psychology Implicit Association Test to identify and quantify biases in a Word Embeding (WE) of the Portuguese language. For this, we use a GloVe model trained on an Internet corpus collection. The results show that several common sense and gender stereotypes can be found in WE. Within the context of professions, we note a historical sexism, since the identified bias often reflects the statistics of gender performance in occupation groups in Brazil. The results show discrimination similar to those of international studies and allow discussing the impact of the use of language models in our society
AB - This work reports experiments based on the Psychology Implicit Association Test to identify and quantify biases in a Word Embeding (WE) of the Portuguese language. For this, we use a GloVe model trained on an Internet corpus collection. The results show that several common sense and gender stereotypes can be found in WE. Within the context of professions, we note a historical sexism, since the identified bias often reflects the statistics of gender performance in occupation groups in Brazil. The results show discrimination similar to those of international studies and allow discussing the impact of the use of language models in our society
KW - Informatics
KW - Artificial Intelligence
KW - Natural Language Processing
KW - Word Embedding
KW - Implicit Association
KW - Gender bias
KW - Discrimination analysis
U2 - 10.5753/stil.2023.233845
DO - 10.5753/stil.2023.233845
M3 - Aufsätze in Konferenzbänden
SP - 53
EP - 62
BT - BRAZILIAN SYMPOSIUM IN INFORMATION AND HUMAN LANGUAGE TECHNOLOGY (STIL)
A2 - Caseli, Helena
A2 - Finatto, Maria Jose Bocorny
PB - Sociedade Brasileira de Computação (SBC)
Y2 - 25 September 2023 through 29 September 2023
ER -