Detecting portuguese and english Twitter users’ gender

Detalhes bibliográficos
Autor(a) principal: Vicente, Marco Paulo Fernandes
Data de Publicação: 2015
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10071/11050
Resumo: Existing social networking services provide means for people to communicate and express their feelings in a easy way. Such user generated content contains clues of user’s behaviors and preferences, as well as other metadata information that is now available for scientific research. Twitter, in particular, has become a relevant source for social networking studies, mainly because: it provides a simple way for users to express their feelings, ideas, and opinions; makes the user generated content and associated metadata available to the community; and furthermore provides easy-to-use web interfaces and application programming interfaces (API) to access data. For many studies, the available information about a user is relevant. However, the gender attribute is not provided when creating a Twitter account. The main focus of this study is to infer the users’ gender from other available information. We propose a methodology for gender detection of Twitter users, using unstructured information found on Twitter profile, user generated content, and later using the user’s profile picture. In previous studies, one of the challenges presented was the labor-intensive task of manually labelling datasets. In this study, we propose a method for creating extended labelled datasets in a semi-automatic fashion. With the extended labelled datasets, we associate the users’ textual content with their gender and created gender models, based on the users’ generated content and profile information. We explore supervised and unsupervised classifiers and evaluate the results in both Portuguese and English Twitter user datasets. We obtained an accuracy of 93.2% with English users and an accuracy of 96.9% with Portuguese users. The proposed methodology of our research is language independent, but our focus was given to Portuguese and English users.
id RCAP_1d2337932058fa005a3d3880633de7d8
oai_identifier_str oai:repositorio.iscte-iul.pt:10071/11050
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Detecting portuguese and english Twitter users’ genderText miningGender classificationTwitter userFeature selectionText classificationMineração de textoClassificação de géneroUtilizador TwitterSelecção de atributosClassificação de textoExisting social networking services provide means for people to communicate and express their feelings in a easy way. Such user generated content contains clues of user’s behaviors and preferences, as well as other metadata information that is now available for scientific research. Twitter, in particular, has become a relevant source for social networking studies, mainly because: it provides a simple way for users to express their feelings, ideas, and opinions; makes the user generated content and associated metadata available to the community; and furthermore provides easy-to-use web interfaces and application programming interfaces (API) to access data. For many studies, the available information about a user is relevant. However, the gender attribute is not provided when creating a Twitter account. The main focus of this study is to infer the users’ gender from other available information. We propose a methodology for gender detection of Twitter users, using unstructured information found on Twitter profile, user generated content, and later using the user’s profile picture. In previous studies, one of the challenges presented was the labor-intensive task of manually labelling datasets. In this study, we propose a method for creating extended labelled datasets in a semi-automatic fashion. With the extended labelled datasets, we associate the users’ textual content with their gender and created gender models, based on the users’ generated content and profile information. We explore supervised and unsupervised classifiers and evaluate the results in both Portuguese and English Twitter user datasets. We obtained an accuracy of 93.2% with English users and an accuracy of 96.9% with Portuguese users. The proposed methodology of our research is language independent, but our focus was given to Portuguese and English users.Os serviços de redes sociais existentes proporcionam meios para as pessoas comunicarem e exprimirem os seus sentimentos de uma forma fácil. O conteúdo gerado por estes utilizadores contém indícios dos seus comportamentos e preferências, bem como outros metadados que estão agora disponíveis para investigação científica. O Twitter em particular, tornou-se uma fonte importante para estudos das redes socias, sobretudo porque fornece um modo simples para os utilizadores expressarem os seus sentimentos, ideias e opiniões; disponibiliza o conteúdo gerado pelos utilizadores e os metadados associados à comunidade; e fornece interfaces web e interfaces de programação de aplicações (API) para acesso aos dados de fácil utilização. Para muitos estudos, a informação disponível sobre um utilizador é relevante. No entanto, o atributo de género não é fornecido ao criar uma conta no Twitter. O foco principal deste estudo é inferir o género dos utilizadores através da informação disponível. Propomos uma metodologia para a detecção de género de utilizadores do Twitter, usando informação não estruturada encontrada no perfil do Twitter, no conteúdo gerado pelo utilizador, e mais tarde usando a imagem de perfil do utilizador. Em estudos anteriores, um dos desafios apresentados foi a tarefa de etiquetar manualmente dados, que revelou exigir bastante trabalho. Neste estudo, propomos um método para a criação de conjuntos de dados etiquetados de uma forma semi-automática, utilizando um conjunto de atributos com base na informação não estruturada de perfil. Utilizando os conjuntos de dados etiquetados, associamos conteúdo textual ao seu género e criamos modelos, com base no conteúdo gerado pelos utilizadores, e na informação de perfil. Exploramos classificadores supervisionados e não supervisionados e avaliamos os resultados em ambos os conjuntos de dados de utilizadores Portugueses e Ingleses do Twitter. Obtivemos uma precisão de 93,2% com utilizadores Ingleses e uma precisão de 96,9% com utilizadores Portugueses. A metodologia proposta é independente do idioma, mas o foco foi dado a utilizadores Portugueses e Ingleses.2016-03-10T12:42:06Z2015-01-01T00:00:00Z20152015-10info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfapplication/octet-streamhttp://hdl.handle.net/10071/11050TID:201080168engVicente, Marco Paulo Fernandesinfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-09T17:38:00Zoai:repositorio.iscte-iul.pt:10071/11050Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T22:17:22.846988Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Detecting portuguese and english Twitter users’ gender
title Detecting portuguese and english Twitter users’ gender
spellingShingle Detecting portuguese and english Twitter users’ gender
Vicente, Marco Paulo Fernandes
Text mining
Gender classification
Twitter user
Feature selection
Text classification
Mineração de texto
Classificação de género
Utilizador Twitter
Selecção de atributos
Classificação de texto
title_short Detecting portuguese and english Twitter users’ gender
title_full Detecting portuguese and english Twitter users’ gender
title_fullStr Detecting portuguese and english Twitter users’ gender
title_full_unstemmed Detecting portuguese and english Twitter users’ gender
title_sort Detecting portuguese and english Twitter users’ gender
author Vicente, Marco Paulo Fernandes
author_facet Vicente, Marco Paulo Fernandes
author_role author
dc.contributor.author.fl_str_mv Vicente, Marco Paulo Fernandes
dc.subject.por.fl_str_mv Text mining
Gender classification
Twitter user
Feature selection
Text classification
Mineração de texto
Classificação de género
Utilizador Twitter
Selecção de atributos
Classificação de texto
topic Text mining
Gender classification
Twitter user
Feature selection
Text classification
Mineração de texto
Classificação de género
Utilizador Twitter
Selecção de atributos
Classificação de texto
description Existing social networking services provide means for people to communicate and express their feelings in a easy way. Such user generated content contains clues of user’s behaviors and preferences, as well as other metadata information that is now available for scientific research. Twitter, in particular, has become a relevant source for social networking studies, mainly because: it provides a simple way for users to express their feelings, ideas, and opinions; makes the user generated content and associated metadata available to the community; and furthermore provides easy-to-use web interfaces and application programming interfaces (API) to access data. For many studies, the available information about a user is relevant. However, the gender attribute is not provided when creating a Twitter account. The main focus of this study is to infer the users’ gender from other available information. We propose a methodology for gender detection of Twitter users, using unstructured information found on Twitter profile, user generated content, and later using the user’s profile picture. In previous studies, one of the challenges presented was the labor-intensive task of manually labelling datasets. In this study, we propose a method for creating extended labelled datasets in a semi-automatic fashion. With the extended labelled datasets, we associate the users’ textual content with their gender and created gender models, based on the users’ generated content and profile information. We explore supervised and unsupervised classifiers and evaluate the results in both Portuguese and English Twitter user datasets. We obtained an accuracy of 93.2% with English users and an accuracy of 96.9% with Portuguese users. The proposed methodology of our research is language independent, but our focus was given to Portuguese and English users.
publishDate 2015
dc.date.none.fl_str_mv 2015-01-01T00:00:00Z
2015
2015-10
2016-03-10T12:42:06Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10071/11050
TID:201080168
url http://hdl.handle.net/10071/11050
identifier_str_mv TID:201080168
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/octet-stream
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799134732899844096