Currently, data analysis is a challenging task, especially in the field of education, because in-depth research is carried out to know, understand and manage the diversity of students who enter for higher education institution and with it, to propose educational strategies to improve the teaching-learning model. The objective of this article was to characterize the pro|44file of admitted student of a public university with respect to their socio-demographic, economic and academic performance variables using K-prototypes algorithm. For this purpose, data from admitted, the entrant's file and their school certificate. It was possible to determine that admitted student fits with 5 profiles, each one with its own characteristics, allowing the grouping of students with similar characteristics, contributing to the improvement of support policies, promoting changes in favor of educational quali...