Minería de datos aplicada en la gestión de la deserción estudiantil en la Universidad Estatal de Bolívar

Currently education is the fundamental pillar for a society in the search for development and welfare, social equity, competitiveness and productivity, it is important to conduct a study of student desertion at the State University of Bolivar because it is not known an exact number of desertions, these can be social, psychological, economic, academic, among others. In different careers, it is possible to verify a quite considerable dropout of subjects/courses, scarcity of economic resources, vocational disorientation, multiple socio-cultural components, poor quality teaching and problems in academic performance, which are shown under different limits, according to the situations that arise in the social and cultural environment of each individual. The objective of this work is to provide a web application to visualize and process data from different SI@NET databases, with the support of the Power BI tool, and ETL processes to transform the information necessary for the analysis, processing and presentation of the results of university dropouts to allow decision making and mitigate this problem. In the present project to determine the indexes and factors of desertion, data mining techniques were used, which are based on a supervised learning model, for the graphical representation scatter plots were used, elevation, regression among others, which allow automating the monitoring of university desertion, and the methodology used CRISP-DM, which serves as a specific basis for the discovery of knowledge of the data, following a series of guidelines in order to exercise a correct development of the project studied.

• Hoofdauteur: Coloma Garófalo, Jesús Antonio (author)
• Formaat: bachelorThesis
• Taal: spa
• Gepubliceerd in: 2022
• Onderwerpen: Minería de datos; Power BI; Deserción estudiantil; Procesos ETL; CRISP-DM
• Online toegang: https://dspace.ueb.edu.ec/handle/123456789/4635