Intent Classifier Model using Recurrent Neural Networks: a case study to help students with doubts about the functionalities of AVA MOODLE

Jeferson Oliveira, Regina Barwaldt, Luiz Oscar Homann de Topin, Joelson Sartori

Resumo


This survey features an intelligent IBM Watson chatbot to improve student participation in Moodle. Recurrent Neural Networks were used to generate a model and classify the intentions of 48 students who interacted with the assistant. A neural network was trained and submitted to the evaluation criteria and obtained 99.80% precision tests. The virtual assistant also demonstrated the ability to collaborate in student service activities. Infrequently asked questions about the use of Moodle. This study showed evidence that it is feasible to implement the chatbot to help with doubts in these contexts.


Palavras-chave


Assistente virtual, IBM Watson, Ambiente Virtual de Aprendizagem, Redes Neurais Recorrentes.

Texto completo:

PDF


DOI: https://doi.org/10.22456/1679-1916.110230

Direitos autorais 2021 RENOTE



RENOTE - Revista Novas Tecnologias na Educação      ISSN 1679-1916

A RENOTE, possui a nota Qualis B1 (Ensino) e B1 (Interdisciplinar) no Sistema de Avaliação da CAPES.

Centro Interdisciplinar de Novas Tecnologias na Educação (CINTED) - Universidade Federal do Rio Grande do Sul (UFRGS)

Avenida Paulo Gama, 110 - Anexo III – 3o andar
Porto Alegre, RS, Brasil – CEP: 90040-060

Telefone: (51) 3308-4100
E-mail: renote@cinted.ufrgs.br