Regresar

Machine learning applied to the analysis of the COVID-19 confinement and its relationship with the performance of higher education students

Abstract:

This article analyzes the academic performance of third level students and the effect of the COVID-19 pandemic due to the change in study modality. Using the Lesmeister methodology, the k-means and hierarchical clustering algorithms were applied to a sample of 400 records corresponding to two study periods, pre-pandemic and in pandemic, of the students of the “Instituto Superior Tecnológico Sudamericano” in the city of Loja, obtaining three groupings related to academic performance. Regarding a general analysis of the groups, it can be established that no significant difference was found in the variables of gender, ethnicity, type of school, employment status, scholarship holders, total household income and household members. However, in the C2 cluster, the students who are mostly married and do not live in the city of Loja lowered their performance during the pandemic due to the study conditions.