idAuthority : um modelo para a identificação de autoridade cognitiva em redes sociais

• Ano de defesa: 2015
• Autor(a) principal: Toledo, Douglas Franscisquini
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual de Maringá; Brasil; Departamento de Informática; Programa de Pós-Graduação em Ciência da Computação; UEM; Maringá, PR; Centro de Tecnologia
• Palavras-chave em Português: Computação; Autoridade cognitiva; Modelo; Redes Sociais; Análise de sentimento; Recomendação de pessoas; Brasil.; Cognitive authority; Social networks; People recommendation; Brazil.; Ciências Exatas e da Terra; Ciência da Computação
• Link de acesso: http://repositorio.uem.br:8080/jspui/handle/1/2527

With the great amount of information available on the web, finding relevant and quality information has become a challenging and complex activity. The popularity of online social networks has contributed to the appearance of new search models that consider the relationship and interaction between people as a means to support information search and recommendation. According to the Cognitive Authority theory, people often resort to other people they trust, or attribute some credibility (i.e., their cognitive authorities) when they need information or knowledge that they do not possess themselves. Adopting the principle that a cognitive authority produces and shares information of better quality and relevance for those who recognizes that authority, we propose the idAuthority: a model for identifying cognitive authorities in social networks that considers and articulates three perspectives for the identification of authorities: 1. Explicit Authority, 2. Implicit Authority and 3. Self-Expertise. The purpose of the idAthority model is to identify and to recommend people who can to contribute to the information need from others people in a particular social network. In order to verify if these recommendations really make sense to people, the idAuthority was instatiated and evaluated in the specific context of a social network called "All We Network", in which the users are teachers that work in the Inclusive Education field in Brazil. The instance used the relationship and interaction between users of the network, and the information published by them to meet the three perspectives of the model. The evaluation was performed on this instance with a group of six teachers who use the TNR to support collaborative discussion of problems in their professional activities. The results obtained by the evaluation showed that the recommendations generated by idAuthority are useful to teachers, indicating that the model is promising to support the recommendation of authorities in social networks.