Publicação

DEVELOPMENT OF AN EDUCATIONAL PLUGIN WITH GENERATIVE AI: A CASE STUDY IN MOODLE

Ver documento

Detalhes bibliográficos
Resumo:This article presents the development and implementation of an educational plugin for the Moodle platform that integrates generative artificial intelligence through Retrieval-Augmented Generation (RAG). The developed system consists of a Moodle block that provides a conversational assistant, chatbot, based on the gemma3:12B model, complemented by gamification elements to increase student engagement. The solution uses the Flowise platform to orchestrate AI flows and implements a progressive reward system based on the number of interactions. The results prove the technical feasibility of integrating open source language models into institutional educational environments, providing an affordable and customizable alternative to commercial assistants. The modular architecture developed allows for scalability and future functional extensions, contributing to the democratization of access to AI-based educational technologies.
Autores principais:Queirós, Ricardo
Outros Autores:Soares, David
Assunto:Moodle Inteligência artificial RAG Gamificação Assistente educativo Moodle Artificial Intelligence RAG Gamification Educational assistant Moodle Inteligencia Artificial RAG Gamifciación Asistente educativo
Ano:2025
País:Portugal
Tipo de documento:artigo
Tipo de acesso:unknown
Instituição associada:Instituto Politécnico do Porto
Idioma:português
Origem:PRATICA - Revista de Pesquisa Multimídia sobre Inovação Pedagógica e Práticas de e-Learning
Descrição
Resumo:This article presents the development and implementation of an educational plugin for the Moodle platform that integrates generative artificial intelligence through Retrieval-Augmented Generation (RAG). The developed system consists of a Moodle block that provides a conversational assistant, chatbot, based on the gemma3:12B model, complemented by gamification elements to increase student engagement. The solution uses the Flowise platform to orchestrate AI flows and implements a progressive reward system based on the number of interactions. The results prove the technical feasibility of integrating open source language models into institutional educational environments, providing an affordable and customizable alternative to commercial assistants. The modular architecture developed allows for scalability and future functional extensions, contributing to the democratization of access to AI-based educational technologies.