Abstract:

Academic desertion within the university system constitutes a challenge for Higher Education Institutions inside and outside the country, as a consequence of social, individual and institutional factors closely related to each other.

Consequently, the study of early abandonment or cessation has been widely recognized as a serious problem, especially at the university level. Therefore, the ability to predict a student's performance could be useful.

With this idea, it is intended to use data mining for the early detection of students at risk of dropping out, analyzing information from students about: academic tutorials, grades, socio-economic data, among others, stored in databases, focusing on finding factors of the behavior of drop out.

The aim is to create a data warehouse that will allow students to clean and transform their data, with the purpose of creating a database with information regarding student behavior factors in order to analyze them.

Next, a data mining model based on the data warehouse will be created to identify possible characteristics and patterns that cause repetition and desertion, and thus better guide students with a higher risk of failure or abandonment.

For this, the CRISP-DM methodology will be applied, which provides a standardized description of the life cycle of a standard project of data analysis, through phases that link the tasks and relationships between them, for the elaboration of models.

In addition, it is intended to use business intelligence to facilitate decision making, through the use of data mining tools, to assist authorities and administrative personnel in the accreditation processes, resource management, and teacher recruitment.

This article presents an outline of the process to follow to obtain the data mining model and thus contribute to future researches.

Keywords:

Evaluating research, educational data mining, educational analytics, learning analytics.