Calculating the influence of tagging people on sentiment analysis

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.creatorRamos, Breno Ladeira-
Autor(es): dc.creatorLasmar, Eduardo-
Autor(es): dc.creatorRosa, Renata Lopes-
Autor(es): dc.creatorRodriguez, Demostenes Zegarra-
Autor(es): dc.creatorGrutzman, Andre-
Data de aceite: dc.date.accessioned2026-02-09T11:23:36Z-
Data de disponibilização: dc.date.available2026-02-09T11:23:36Z-
Data de envio: dc.date.issued2019-08-22-
Data de envio: dc.date.issued2019-08-22-
Data de envio: dc.date.issued2018-
Fonte completa do material: dc.identifierhttps://repositorio.ufla.br/handle/1/36401-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/capes/1139905-
Descrição: dc.descriptionSocial media makes it possible for anyone to share and transmit their opinions and sentiments to the rest of the world via the Internet. Recently, sentiment analysis is being used to investigate the opinions posted on Online Social Networks (OSN) based on multiple inputs, such as user profile characteristics, slang, emoticons, among others. However, the current sentiment analysis tools do not consider the influence of tagging people on OSN. In this context, this paper analyzes the impact of the tagging parameter on the global sentiment score of a text. The experimental results of subjective tests show that a correction factor must be considered in case of tagging people. Experimental results demonstrate that the tagging parameter affects the sentiment intensity value, in different ways, depending on the gender of the person who wrote the text and the sentiment polarity of the text. The new sentiment intensity metric considering the tagging people parameter reaches a Pearson Correlation Coefficient of 0.93 and a maximum error of 0.07, for texts of negative polarity written by women, at a 5-point scale. Furthermore, a mobile application with the new sentiment metric, which considers the tagging parameter, is built.-
Descrição: dc.descriptionhttps://ieeexplore.ieee.org/document/8555772/authors#authors-
Idioma: dc.languageen-
Direitos: dc.rightsrestrictAccess-
Palavras-chave: dc.subjectSentiment analysis-
Palavras-chave: dc.subjectOnline social network-
Palavras-chave: dc.subjectData mining-
Palavras-chave: dc.subjectTagging people-
Título: dc.titleCalculating the influence of tagging people on sentiment analysis-
Tipo de arquivo: dc.typeTrabalho apresentado em evento-
Aparece nas coleções:Repositório Institucional da Universidade Federal de Lavras (RIUFLA)

Não existem arquivos associados a este item.