| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| DM_CarlosSilva_MEI_2018 | 7.6 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
A adoção das plataformas sociais online como um espaço comum para a virtualização das
identidades está também correlacionada com a replicação de riscos sociais do mundo real no
mundo virtual. Com este aumento da presença online, alguns problemas como o bullying
começam-se a intensificar. Cyberbullying é uma prática muito comum entre os jovens,
tornando-se muito mais presente devido à sua constante atividade online, especialmente
através das redes sociais. Este trabalho tem como objetivo apresentar uma proposta que
combine técnicas de análise textual, visual e cognitiva sobre o conteúdo que seja
disponibilizado online, de modo a ter um sistema inteligente e autónomo que consiga
identificar e classificar uma possível situação de bullying. Pretende-se analisar o texto de uma
publicação online, verificar a presença de pessoas em conteúdo multimédia e identificar os
indivíduos envolvidos num possível caso de bullying, procurando mitigar futuras situações
semelhantes.
The adoption of online social platforms as a common space for the virtualisation of identities is also correlated with the replication of real-world social hazards in the virtual world. With this increase in online presence, some problems like bullying begin to intensify. Cyberbullying is a very common practice among young people nowadays, becoming much more present due to their constant online activity, especially in online social networks This work aims to present a proposal that combines textual, visual and cognitive analysis techniques on the content that is available online, in order to have an intelligent and autonomous system that can identify and classify a possible bullying situation. It intends to analyze the text of an online publication, verify the presence of people in multimedia content and identify individuals involved in a possible case of bullying, trying to mitigate future similar situations.
The adoption of online social platforms as a common space for the virtualisation of identities is also correlated with the replication of real-world social hazards in the virtual world. With this increase in online presence, some problems like bullying begin to intensify. Cyberbullying is a very common practice among young people nowadays, becoming much more present due to their constant online activity, especially in online social networks This work aims to present a proposal that combines textual, visual and cognitive analysis techniques on the content that is available online, in order to have an intelligent and autonomous system that can identify and classify a possible bullying situation. It intends to analyze the text of an online publication, verify the presence of people in multimedia content and identify individuals involved in a possible case of bullying, trying to mitigate future similar situations.
Description
Dissertação de Mestrado em Engenharia Informática
Keywords
Cyberbullying Machine Learning Redes Sociais Análise Textual Análise Visual Análise Cognitiva
Pedagogical Context
Citation
Publisher
Instituto Politécnico do Porto. Escola Superior de Tecnologia e Gestão