Application of an Emotion Miner in Discussion Forums of a Massive Open Online Course (MOOC) in Brazil: an Approach Using the Naive Bayes Algorithm

• Autor(a) principal: Souza, Vanessa Faria de
• Data de Publicação: 2022
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: EAD em Foco
• Texto Completo: https://eademfoco.cecierj.edu.br/index.php/Revista/article/view/1732

MOOCs are gradually evolving which is due to the wide dissemination of virtual learning environments, which provide means of interaction for participants, one of which is the discussion forum, which has a lot of information about student engagement. However, reading all the posts is a difficult task, as MOOCs tend to have a very high number of students enrolled. In this sense, text mining can help teachers gain relevant knowledge about students' posts. Thus, in this study, an emotion miner was implemented for MOOC forums, using the Python programming language, in order to identify and analyze the feelings that each student expresses when interacting with others, in these environments. The results obtained, in initial experiments, show that the miner proved to be efficient in extracting students' emotions, reaching an accuracy of 40% and that positive feelings such as joy and surprise reflect on the conclusion of the MOOCs, while negative feelings such as sadness and anger are indicative of dropping out of the course. Keywords: MOOCs. Discussion forums. Miner of emotions. Naive Bayes Algorithm.