Loading...
Publication
Sistema de recomendaĆ§Ć£o no contexto educacional
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 8.99 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Com a constante evoluĆ§Ć£o tecnolĆ³gica, as escolas portuguesas tĆŖm vindo a transitar progressivamente
dos manuais escolares em papel para versƵes digitais. Esta transiĆ§Ć£o exige
uma crescente adaptaĆ§Ć£o dos alunos ao uso de ferramentas tecnolĆ³gicas no seu processo
de aprendizagem. No entanto, esta transiĆ§Ć£o nem sempre Ć© acompanhada por soluƧƵes
eficazes que respondam Ć s necessidades especĆficas de cada estudante, tornando essencial
o desenvolvimento de plataformas que promovam a personalizaĆ§Ć£o do ensino.
Neste enquadramento, o presente projeto teve como objetivo principal investigar de que
forma um sistema de recomendaĆ§Ć£o personalizado pode apoiar e facilitar o processo de
aprendizagem no contexto do ensino escolar. Para tal, procedeu-se Ć realizaĆ§Ć£o de uma
revisĆ£o do estado da arte sobre sistemas de recomendaĆ§Ć£o aplicados ao ensino. Adicionalmente,
procurou-se explorar o potencial dos Large Language Models (LLMs) em contextos
educativos e analisar criticamente as plataformas educacionais existentes.
A metodologia adotada envolveu uma abordagem exploratĆ³ria, sustentada por pesquisa bibliogrĆ”fica
em bases cientĆficas. Na seleĆ§Ć£o dos artigos, consideraram-se como critĆ©rios
critĆ©rios a atualidade, a relevĆ¢ncia e o foco na aplicaĆ§Ć£o de sistemas de recomendaĆ§Ć£o em
contextos educacionais.
No Ć¢mbito do projeto, foi desenvolvida uma plataforma educacional que disponibiliza conteĆŗdos
educativos, integrando um sistema de recomendaĆ§Ć£o personalizado, um assistente
virtual de apoio ao estudo entre outras funcionalidades complementares.
Os resultados obtidos demonstraram que a plataforma desenvolvida foi bem recebida pelos
participantes. Estes utilizadores beneficiaram de recomendaƧƵes personalizadas de conteĆŗdo
e de um assistente virtual capaz de apoiar o estudo atravƩs de comandos personalizados. A
anĆ”lise das mĆ©tricas indicou um bom desempenho do sistema de recomendaĆ§Ć£o, revelando
a sua capacidade de adaptaĆ§Ć£o aos interesses e dificuldades dos estudantes.
Conclui-se que uma plataforma educacional gratuita, equipada com funcionalidades inteligentes
e personalizadas, pode representar uma mais-valia significativa no apoio ao ensino
digital. Com base nos dados recolhidos, foram ainda implementadas melhorias e adicionadas
novas funcionalidades, contribuindo para o desenvolvimento de futuras soluƧƵes tecnolĆ³gicas
mais eficazes e centradas no aluno.
With the constant technological evolution, Portuguese schools have been progressively transitioning from paper textbooks to digital versions. This transition requires increasing adaptation from students to the use of technological tools in their learning process. However, this transition is not always accompanied by effective solutions that respond to the specific needs of each student, making it essential to develop platforms that promote personalized teaching. In this context, the main objective of this project was to investigate how a personalized recommendation system can support and facilitate the learning process in the school education context. To this end, a state-of-the-art review on recommendation systems applied to education was conducted. Additionally, the potential of Large Language Models (LLMs) in educational contexts was explored and existing educational platforms were critically analyzed. The adopted methodology involved an exploratory approach, supported by bibliographic research in scientific databases. In the selection of articles, the criteria considered were timeliness, relevance, and focus on the application of recommendation systems in educational contexts. Within the scope of the project, an educational platform was developed that provides educational content, integrating a personalized recommendation system, a virtual study support assistant, and other complementary functionalities. The results obtained demonstrated that the developed platform was well received by the participants. These users benefited from personalized content recommendations and a virtual assistant capable of supporting study through personalized commands. The analysis of metrics indicated good performance of the recommendation system, revealing its ability to adapt to studentsā interests and difficulties. It is concluded that a free educational platform, equipped with intelligent and personalized functionalities, can represent a significant added value in supporting digital education. Based on the collected data, improvements were implemented and new functionalities were added, contributing to the development of more effective and student-centered future technological solutions.
With the constant technological evolution, Portuguese schools have been progressively transitioning from paper textbooks to digital versions. This transition requires increasing adaptation from students to the use of technological tools in their learning process. However, this transition is not always accompanied by effective solutions that respond to the specific needs of each student, making it essential to develop platforms that promote personalized teaching. In this context, the main objective of this project was to investigate how a personalized recommendation system can support and facilitate the learning process in the school education context. To this end, a state-of-the-art review on recommendation systems applied to education was conducted. Additionally, the potential of Large Language Models (LLMs) in educational contexts was explored and existing educational platforms were critically analyzed. The adopted methodology involved an exploratory approach, supported by bibliographic research in scientific databases. In the selection of articles, the criteria considered were timeliness, relevance, and focus on the application of recommendation systems in educational contexts. Within the scope of the project, an educational platform was developed that provides educational content, integrating a personalized recommendation system, a virtual study support assistant, and other complementary functionalities. The results obtained demonstrated that the developed platform was well received by the participants. These users benefited from personalized content recommendations and a virtual assistant capable of supporting study through personalized commands. The analysis of metrics indicated good performance of the recommendation system, revealing its ability to adapt to studentsā interests and difficulties. It is concluded that a free educational platform, equipped with intelligent and personalized functionalities, can represent a significant added value in supporting digital education. Based on the collected data, improvements were implemented and new functionalities were added, contributing to the development of more effective and student-centered future technological solutions.
Description
Keywords
EducaĆ§Ć£o em Portugal Recursos Educacionais Digitais Sistemas de RecomendaĆ§Ć£o Large Language Models (LLM) InteligĆŖncia Artificial no Ensino EducaĆ§Ć£o em Portugal Recursos educacionais digitais Sistemas de recomendaĆ§Ć£o InteligĆŖncia artificial no ensino