Análise da produção científica dos cursos de pós-graduação utilizando redes neurais e modelagem de tópicos

• Ano de defesa: 2020
• Autor(a) principal: Casara, Mary Adriana
• Orientador(a): Notargiacomo, Pollyana Coelho da Silva
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Presbiteriana Mackenzie
• Palavras-chave em Português: indicadores bibliométricos; mineração de textos; mineração de dados; mapas auto-organizáveis; redes neurais artificiais; modelagem de tópicos
• Área do conhecimento CNPq: CNPQ::ENGENHARIAS
• Link de acesso: https://dspace.mackenzie.br/handle/10899/28617

Scientific production is one of the components considered by CAPES in the four-year evaluation of Stricto Sensu Graduate Programs (Masters and Doctoral Programs) in Brazil and directly influences the grade awarded to these Programs. Higher grades imply greater visibility and, consequently, the attraction of financial resources in the form of scholarships and funding for research. Thus, this paper aims to analyze the scientific production represented by the theses, dissertations, projects and book chapters generated in the period from 2013 to 2016, that is, the period coinciding with the 2017 evaluation for the four-year preceding period, in order to understand its relationship with the performance of the original Programs. This work not only consists of the quantitative analysis of the production of Graduate Programs, but also seeks, through techniques of artificial neural networks and text mining, to generate groups of Programs based on the similarity of their productions. The results obtained allow the identification of predominant patterns and characteristics of the Programs considered to be of excellence, which can be used as a reference by other Programs that wish to achieve the same performance.