• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.100.2021.tde-28012022-074813
Document
Author
Full name
Adriano dos Santos Rodrigues da Silva
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2021
Supervisor
Committee
Roman, Norton Trevisan (President)
Carvalho, Ariadne Maria Brito Rizzoni
Pardo, Thiago Alexandre Salgueiro
Title in Portuguese
Estudo de modelos distribucionais para detecção de discurso de ódio em português
Keywords in Portuguese
Discurso de ódio
Mineração de texto
Redes sociais
Abstract in Portuguese
Com o surgimento das redes sociais, os usuários passaram de consumidores a produtores de conteúdo, sendo que qualquer usuário tem a liberdade de emitir sua opinião. Devido à grande quantidade de conteúdo que os usuários publicam nas redes sociais, torna-se impossível que o monitoramento seja feito por agente humano, portanto é necessário encontrar uma forma para que essa supervisão seja de forma automática. Entretanto, esse problema é pouco explorado para o português, sendo que a maioria das pesquisas são dedicadas ao idioma inglês. Além disso, os modelos distribucionais podem ser utilizados em diversas tarefas, inclusive na tarefa de identicação de discurso de ódio em tweets. Nos experimentos realizados nesta pesquisa, esses modelos obtiveram desempenho superior em relação aos métodos tradicionais como N-Gram combinada com SVM
Title in English
Study of distributional models for the detection of hate speech in Portuguese
Keywords in English
Hate speech
Social networks
Text mining
Abstract in English
With the rise of social networks, users have gone from being consumers to content producers, and any user is free to express their opinion. Due to the large amount of content that users publish on social networks, it is impossible for monitoring to be carried out by a human agent, so it is necessary to find a way for this supervision to be automatic. However, this problem is little explored for Portuguese, as most of the research is dedicated to the English language. Furthermore, distributional models can be used in several tasks, including the task of identifying hate speech in tweets. In the experiments performed in this research, these models obtained superior performance compared to traditional methods such as N-Gram combined with SVM
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2022-07-28
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2022. All rights reserved.