Automatic Identification of Learning Styles Through Behavioral Patterns

The learning style is characterized by the preferences that the student is acquiring throughout his life and with the interaction in the environment in which he develops. As a traditional method to identify the learning style, the application of questionnaires is carried out; however, they may present inaccurate results due to lack of interest in answering the questionnaires. Currently, there are automatic methods to identify the learning style as the observation of the student's behavior, while interacting with learning objects in an academic course. Learning objects are those digital tools such as chat rooms, reading materials, exams, among others. These objects allow the actions carried out to be recorded and stored in order to be studied. Therefore, this study presents an evolutionary algorithm to optimize the grouping of students according to their learning style based on the structured learning objects. Thus, the objective is to form groups according to the value of the actions carried out in an academic course and to predict the learning style.