Sistema de recomendação de trajetórias curriculares para discentes do curso de Bacharelado em Engenharia de Computação do ICT-UNIFESP
Data
2022-02-23
Autores
Gomes, Pedro Spoljaric 
Orientadores
Oliveira, Tiago de
Tipo
Trabalho de conclusão de curso
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Para auxiliar os estudantes do curso de Bacharelado em Engenharia de Computação do Instituto de Ciência e Tecnologia da Universidade Federal de São Paulo (ICT - UNIFESP) no planejamento de suas trajetórias acadêmicas, foi desenvolvido um sistema computacional para recomendação de matrizes horárias semestrais personalizadas. Para criar um embasamento teórico sólido sobre este tipo de sistema, foi realizada uma revisão sistemática da literatura nacional com o tema sistemas de recomendação na educação. Em seguida, foi feito um levantamento de requisitos do sistema através de uma pesquisa realizada com 84 estudantes dos cursos de graduação do ICT - UNIFESP, para coleta de informações sobre as dificuldades enfrentadas e experiências vividas por eles em relação à sua trajetória acadêmica. A partir da análise das respostas recebidas, as funcionalidades foram definidas e foi construído e disponibilizado como uma plataforma web um protótipo do sistema computacional. O sistema possui em sua base de dados o catálogo oficial de disciplinas e professores do campus e informações sobre a oferta de unidades curriculares das matrizes semestrais. Usando esses dados, em conjunto com o histórico acadêmico do estudante e suas preferências de disponibilidade, o sistema busca recomendar uma matriz horária a partir de um nível de prioridade estabelecido para cada unidade curricular. Após a implementação do sistema, o mesmo foi disponibilizado para os estudantes para validação através de um teste de usabilidade utilizando o modelo System Usability Scale (SUS), juntamente com um espaço aberto para submissão de sugestões e pontos de melhoria. O teste foi realizado por 14 estudantes e obteve uma pontuação de 92,32 de 100 pontos na escala SUS, o que é considerado como ``melhor imaginável''. Por fim, o nível de prontidão do sistema foi avaliado por meio das escalas Technology Readiness Levels (TRL) e Innovation Readiness Levels (IRL), que apontaram um alto nível de maturidade. Dessa forma, é esperada a possibilidade de uso regular do sistema nos períodos oficiais de rematrícula no futuro, com o objetivo de auxiliar os estudantes em suas trajetórias curriculares.
To help students of the Bachelor's degree in Computer Engineering at the Institute of Science and Technology of the Federal University of São Paulo (ICT - UNIFESP) in the planning of their academic paths, a computer system was developed to recommend semester disciplines in a customized way. In order to create a solid theoretical foundation about this type of system, a systematic review of the national literature was carried out about recommender systems in education. Then, a survey of system requirements was carried out with 84 undergraduate students from ICT - UNIFESP, to collect information about the difficulties faced and experiences lived by them in relation to their academic trajectory. From the analysis of the responses received, the features were defined and a prototype of the computer system was built and made available as a web platform. The system has in its database the official catalog of disciplines and professors of the campus and information about the offer of disciplines on each semester. Using that data, the user's availability preferences and their academic history, the system seeks to recommend an optimal set of disciplines using a priority level established for each of them. After the implementation of the system, it was made available to the students for validation through a usability test using the System Usability Scale (SUS) model, together with an open field for submitting suggestions. The test was performed by 14 students and obtained a score of 92.32 out of 100 points on the SUS scale, which is considered as ``best imaginable''. Finally, the system readiness level was evaluated using the Technology Readiness Levels (TRL) and Innovation Readiness Levels (IRL) scales, which indicated a high level of maturity. In this way, the possibility of regular use of the system in official re-enrollment periods is expected in the future, with the goal of helping students in their curricular trajectories.
To help students of the Bachelor's degree in Computer Engineering at the Institute of Science and Technology of the Federal University of São Paulo (ICT - UNIFESP) in the planning of their academic paths, a computer system was developed to recommend semester disciplines in a customized way. In order to create a solid theoretical foundation about this type of system, a systematic review of the national literature was carried out about recommender systems in education. Then, a survey of system requirements was carried out with 84 undergraduate students from ICT - UNIFESP, to collect information about the difficulties faced and experiences lived by them in relation to their academic trajectory. From the analysis of the responses received, the features were defined and a prototype of the computer system was built and made available as a web platform. The system has in its database the official catalog of disciplines and professors of the campus and information about the offer of disciplines on each semester. Using that data, the user's availability preferences and their academic history, the system seeks to recommend an optimal set of disciplines using a priority level established for each of them. After the implementation of the system, it was made available to the students for validation through a usability test using the System Usability Scale (SUS) model, together with an open field for submitting suggestions. The test was performed by 14 students and obtained a score of 92.32 out of 100 points on the SUS scale, which is considered as ``best imaginable''. Finally, the system readiness level was evaluated using the Technology Readiness Levels (TRL) and Innovation Readiness Levels (IRL) scales, which indicated a high level of maturity. In this way, the possibility of regular use of the system in official re-enrollment periods is expected in the future, with the goal of helping students in their curricular trajectories.