Aplicación del algoritmo K-Medoid para la segmentación de los alumnos ingresantes de una universidad.

Abstract

Currently, in the area within higher education, data management has become essential for academic decision making and the improvement of educational processes. Analytics and statistics have been taken to the technological field, where the processes automation and the large databases management through Machine Learning algorithms are the most used, among which are the clustering algorithms, whose purpose is to group data by similarity. The objective of this study was to find types of university students with respect to their sociodemographic, economic and academic performance variables, using the K-medoid algorithm on data of students entering the Universidad Nacional Agraria La Molina in Lima, Peru. It was determined that the students under study can be segmented into 3 groups, each with its own characteristics, which will make it possibleto promote changes in favor of educational quality and promote the renovation of teaching spaces in a personalized way around the type of student that the university manages.