Uma metodologia para a predição de afinidades e formação de grupos de trabalho a partir de redes sociais

• Ano de defesa: 2014
• Autor(a) principal: Douglas Donizeti de Castilho Braz
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Formação de times; Relacionamento negativo; Teste sociométrico; Redes sociais online; Capital social; Conflito (Psicologia); Relações humanas; Computação; Redes sociais on-line
• Link de acesso: http://hdl.handle.net/1843/ESBF-9KQGV7

Students often have to team up for class projects. In this process, our conjecture is that they group based not only on the performance in the classroom (the grades of the students, for example), but also from their interpersonal relationships (for example, if they trust each other). With best of our knowledge, no studies were found on the relationship between team formation for class projects and social media. In this context, we propose a new methodology to identify work affinities from social media data. The proposed methodology was applied in a group of college students in a given classroom. First, through a questionnaire, we asked students to tell us who they would like to work or not in this room. Then, through an application developed from the Facebook API, we collected data from their online interactions, and finally tested some predictors for team building. Thus, it was concluded that self-organized team members selection can be better estimated from Facebook-derived proxies for the strength of the ties, popularity, extraversion and homophily than proficiency metrics. These results have important implications for the theoretical literature teaming. Currently, companies work with dynamic cooperative, where the harmonic interpersonal relationships is of great interest institutions. Practices for harmonic grouping are used by human resources departments to maximize the performance of work teams. Moreover, there are practical implications for online educational platforms in the performing of collaborative activities.