LEARNING ANALYTICS TO CLASSIFY STUDENTS ACCORDING TO THEIR ACTIVITY IN MOODLE

In the last years, the use of Learning Analytics (LA) tools has been increasing to process the available information about students. In this study, we have analyzed the interactions in the educational scenario collected through Moodle in order to provide a classification of students by using clustering techniques, to improve the learning process. Students online activities generate large quantities of data that before were wasted since no LA tools were available to process them. With the irruption of the big data techniques in the educational sector, a lot of information can be easily treated to extract behaviors of the students and classify them according to their profiles. These analyses define models of action that help both the in-service teacher and the new teacher who joins the department. In addition, detailed analysis of this information may help us in the study of a possible relationship between different indicators of use of these platforms and the performance of students, both generated throughout the learning process as those coming from summative evaluation.

In this study, we provide the construction of a data analysis model facilitated by the Moodle platform, from the different interactions between the teacher, the student and the developed subject; so that this information can be transformed into knowledge and their understanding can help to the improvement of teaching practice. In particular, it addresses the application of these models to the improvement of the students' learning strategies according to their typology.