Comparing Methods for Mitigating Gender Bias in Word Embedding

dc.contributor.authorRonchieri, Elisabetta
dc.contributor.authorBiagi, Clara
dc.date.accessioned2022-12-27T18:55:01Z
dc.date.available2022-12-27T18:55:01Z
dc.date.issued2023-01-03
dc.description.abstractWord embedding captures the semantic and syntactic meaning of words into dense vectors. It contains biases learning from data that include constructs, cultural stereotypes, and inequalities of the society. Many methods for removing bias in traditional word embedding have been proposed. In this study we use the original GloVe word embedding and perform a comparison among debiasing methods built on top of GloVe in order to determine which methods perform the best removing bias. We have defined half-sibling regression, repulsion attraction neutralization GloVe method and compared it with gender-preserving, gender-neutral GloVe method and other debiased methods. According to our results, no methods outperform in all the analyses and completely remove gender information from gender neutral words. Furthermore, all the debiasing methods perform better than the original GloVe.
dc.format.extent10
dc.identifier.doi10.24251/HICSS.2023.091
dc.identifier.isbn978-0-9981331-6-4
dc.identifier.urihttps://hdl.handle.net/10125/102720
dc.language.isoeng
dc.relation.ispartofProceedings of the 56th Hawaii International Conference on System Sciences
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectAccountability, Evaluation, and Obscurity of AI Algorithms
dc.subjectgender bias
dc.subjectglove
dc.subjectnatural language processing
dc.subjectword embedding
dc.titleComparing Methods for Mitigating Gender Bias in Word Embedding
dc.type.dcmitext
prism.startingpage722

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
0070.pdf
Size:
1.98 MB
Format:
Adobe Portable Document Format