Educational data mining application for improving the academic tutorial sessions, and the reduction of early dropout in undergraduate students

This paper is part of a research related to the application of educational data mining and processes derived from learning and academic analytics. The aim is to identify situations of early dropout risk in undergraduate students and to be able to act proactively through academic tutoring sessions. First-year university students must adapt to a new centre and new teaching methodologies, in numerous cases derived from situations such as the current global pandemic. Therefore, adapting to the new situation and the changes in their study habits that it implies are essential in order to successfully complete and successfully adjust to the course. This article carries out an analytical study of the probability of students dropping out based on various samples collected in the first semester in order to detect at an early stage those students who are likely to have difficulties in passing the course or who could drop out. To corroborate the results, they have been compared with the perception of the tutors of each of the students surveyed in order to be able to act proactively. A prediction was made from the relationship established between the variables studied of the percentage of newly enrolled students who will have no problems in their studies, those who will drop out in the first year and those who have a high probability of dropping out. Tutors will have to carry out an exhaustive monitoring of this entire segment through the application of motivation and support techniques that will enable them to redirect their choices.