Inferência de personalidade a partir de textos de rede social utilizando um léxico afetivo em português brasileiro

• Ano de defesa: 2016
• Autor(a) principal: Machado, Antonio Aliberte de Andrade
• Orientador(a): Nunes, Maria Augusta Silveira Netto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Sergipe
• Programa de Pós-Graduação: Pós-Graduação em Ciência da Computação
• País: Brasil
• Palavras-chave em Português: Computação; Personalidade; Redes sociais on-line; Facebook; Lexicologia; Facetas; Big Five
• Palavras-chave em Inglês: Léxico; Personality; Facets; Lexicon
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://ri.ufs.br/handle/riufs/3375

This máster thesis presents research on the correlation of lexical information in texts in Portuguese with personality characteristics and model Big Five facets of IPIP-NEO. It elaborates, especially, on the use of classes of affective feelings LIWC lexicon. The main goal of this work is to relate the factors of the Big Five model and the IPIP-NEO facets of IPIP-NEO 120 and TIPI questionnaires with the posts of Facebook social network. For this, a tool called Personalitatem Lexicon was built. The methodology used to achieve the research objectives was bibliographical which were researched and analyzed the work done on personality inferences from texts. The result of the experiment shows that the inference of personality from the questionnaires have more precise conclusions for the same contain specific questions and answers to measure such characteristic. Already personality inference for posts is more susceptible to noise because not all experienced situations are exposed on the social network. However, inference of personality posts is possible, but the results are the moments