Cargando…

Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content

In this paper we present a benchmark dataset generated as part of a project for automatic identification of misogyny within online content, which focuses in particular on memes. The benchmark here described is composed of 800 memes collected from the most popular social media platforms, such as Face...

Descripción completa

Detalles Bibliográficos
Autores principales: Gasparini, Francesca, Rizzi, Giulia, Saibene, Aurora, Fersini, Elisabetta
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9471366/
https://www.ncbi.nlm.nih.gov/pubmed/36117643
http://dx.doi.org/10.1016/j.dib.2022.108526
_version_ 1784789057913487360
author Gasparini, Francesca
Rizzi, Giulia
Saibene, Aurora
Fersini, Elisabetta
author_facet Gasparini, Francesca
Rizzi, Giulia
Saibene, Aurora
Fersini, Elisabetta
author_sort Gasparini, Francesca
collection PubMed
description In this paper we present a benchmark dataset generated as part of a project for automatic identification of misogyny within online content, which focuses in particular on memes. The benchmark here described is composed of 800 memes collected from the most popular social media platforms, such as Facebook, Twitter, Instagram and Reddit, and consulting websites dedicated to collection and creation of memes. To gather misogynistic memes, specific keywords that refer to misogynistic content have been considered as search criterion, considering different manifestations of hatred against women, such as body shaming, stereotyping, objectification and violence. In parallel, memes with no misogynist content have been manually downloaded from the same web sources. Among all the collected memes, three domain experts have selected a dataset of 800 memes equally balanced between misogynistic and non-misogynistic ones. This dataset has been validated through a crowdsourcing platform, involving 60 subjects for the labelling process, in order to collect three evaluations for each instance. Two further binary labels have been collected from both the experts and the crowdsourcing platform, for memes evaluated as misogynistic, concerning aggressiveness and irony. Finally for each meme, the text has been manually transcribed. The dataset provided is thus composed of the 800 memes, the labels given by the experts and those obtained by the crowdsourcing validation, and the transcribed texts. This data can be used to approach the problem of automatic detection of misogynistic content on the Web relying on both textual and visual cues, facing phenomenons that are growing every day such as cybersexism and technology-facilitated violence.
format Online
Article
Text
id pubmed-9471366
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Elsevier
record_format MEDLINE/PubMed
spelling pubmed-94713662022-09-15 Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content Gasparini, Francesca Rizzi, Giulia Saibene, Aurora Fersini, Elisabetta Data Brief Data Article In this paper we present a benchmark dataset generated as part of a project for automatic identification of misogyny within online content, which focuses in particular on memes. The benchmark here described is composed of 800 memes collected from the most popular social media platforms, such as Facebook, Twitter, Instagram and Reddit, and consulting websites dedicated to collection and creation of memes. To gather misogynistic memes, specific keywords that refer to misogynistic content have been considered as search criterion, considering different manifestations of hatred against women, such as body shaming, stereotyping, objectification and violence. In parallel, memes with no misogynist content have been manually downloaded from the same web sources. Among all the collected memes, three domain experts have selected a dataset of 800 memes equally balanced between misogynistic and non-misogynistic ones. This dataset has been validated through a crowdsourcing platform, involving 60 subjects for the labelling process, in order to collect three evaluations for each instance. Two further binary labels have been collected from both the experts and the crowdsourcing platform, for memes evaluated as misogynistic, concerning aggressiveness and irony. Finally for each meme, the text has been manually transcribed. The dataset provided is thus composed of the 800 memes, the labels given by the experts and those obtained by the crowdsourcing validation, and the transcribed texts. This data can be used to approach the problem of automatic detection of misogynistic content on the Web relying on both textual and visual cues, facing phenomenons that are growing every day such as cybersexism and technology-facilitated violence. Elsevier 2022-08-20 /pmc/articles/PMC9471366/ /pubmed/36117643 http://dx.doi.org/10.1016/j.dib.2022.108526 Text en © 2022 Published by Elsevier Inc. https://creativecommons.org/licenses/by/4.0/This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Data Article
Gasparini, Francesca
Rizzi, Giulia
Saibene, Aurora
Fersini, Elisabetta
Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
title Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
title_full Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
title_fullStr Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
title_full_unstemmed Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
title_short Benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
title_sort benchmark dataset of memes with text transcriptions for automatic detection of multi-modal misogynistic content
topic Data Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9471366/
https://www.ncbi.nlm.nih.gov/pubmed/36117643
http://dx.doi.org/10.1016/j.dib.2022.108526
work_keys_str_mv AT gasparinifrancesca benchmarkdatasetofmemeswithtexttranscriptionsforautomaticdetectionofmultimodalmisogynisticcontent
AT rizzigiulia benchmarkdatasetofmemeswithtexttranscriptionsforautomaticdetectionofmultimodalmisogynisticcontent
AT saibeneaurora benchmarkdatasetofmemeswithtexttranscriptionsforautomaticdetectionofmultimodalmisogynisticcontent
AT fersinielisabetta benchmarkdatasetofmemeswithtexttranscriptionsforautomaticdetectionofmultimodalmisogynisticcontent