Aplicação de técnicas de Multimodal Learning Analytics para identificação de estados emocionais durante atividades de pensamento computacional.

Abstract

With the rapid spread of digital technologies, we see a growth in the construction and adoption of new educational technologies, these technologies that can be used to assist and even improve the teaching-learning process. With the adoption of Distance Education, teachers lose control over an individual analysis of each student, since within a virtual environment there are no limits to the number of students. Furthermore, there is no way to analyze each student in a collaborative task, which corresponds to their learning within an activity. That said, the objective of this work is to use Multimodal Learning Analitycs techniques in order to analyze the flow of emotional states between pairs of students, along three activities related to the development of Computational Thinking skills, in such a way that through from the understanding of the students’ emotional states throughout the activity, we can get a sense of their profile throughout the performance of the work. For this, we collected data from two pairs of high school students, with different collaboration profiles and knowledge about Computational Thinking, in a practical collaborative task on the Scratch platform, with activities corresponding to the development of skills such as algorithmic thinking, abstraction, decomposition and pattern recognition. The activities were carried out through the Google Meet and with the record in video of the same platform, we were able to use the specific algorithms, available on the EZMMLA platform, to capture emotional states throughout the activity and with their analysis extract useful information to assist in the