Análise de dados e de sentimentos para melhoria de resultados em Sistemas de Avaliação Institucional

• Ano de defesa: 2024
• Autor(a) principal: Gustavo Kataoka
• Orientador(a): Amaury Antonio de Castro Junior
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Fundação Universidade Federal de Mato Grosso do Sul
• País: Brasil
• Palavras-chave em Português: Sistemas de Avaliação Institucional
• Link de acesso: https://repositorio.ufms.br/handle/123456789/8647

This work aims to develop improvements in the Institutional Assessment System of the Federal University of Mato Grosso do Sul. To this end, a survey was carried out among its users with the aim of discovering the needs and opportunities for improvement to be implemented in the system. The survey revealed two needs, the first being the automation of a report that is carried out using a spreadsheet, and the second the implementation of a classifier of textual responses provided by system respondents. In addition to the meeting with users, a systematic literature review was carried out to investigate whether the desired improvements were common to other institutions and what techniques and methods were used to resolve them. The bibliographical research revealed that, in textual analysis, some institutions use artificial intelligence and machine learning techniques. This work, in turn, presents, as an improvement to the system, the development of the Action Report and the sentiment analyzer using BERT. A Time Series report discovered in RSL was included as an improvement developed in this work. With the implementation of these improvements in the system, it is expected to make the work routine of its users more efficient and productive, helping them in decision making.