Rediseño de la herramienta predictiva para la proyección de estudiantes universitarios por asignatura y periodo

In the field of higher education, a growing challenge has been identified in the management of course offerings during the most recent semesters. Students and school administration face ongoing hardship with the availability and scheduling of courses, which in result affects the academic experience and generates additional costs to the institution. This project focuses on addressing the lack of accuracy for predicting the number of students who will enroll each semester in any given course offered. The main goal of this work is to improve academic planning by redesigning a more accurate predictive model. To achieve this, a methodology based on the analysis of historical enrollment data, the application of advanced predictive modeling techniques and the inclusion of a new variable based on the estimated number of failed students per course will be used. This approach will allow for a more accurate prediction of student demand on different courses, thus reducing the need for last-minute adjustments and optimizing resource allocation.

• Autores: Escobar Barranco, Edgardo Mario; Rivero Guzmán, Anderson; Fonseca Aguinaga, Dubbys Esther
• Fecha de publicación: 2024
• Institución: Universidad del Norte
• Repositorio: Repositorio Uninorte
• Idioma: eng
• OAI Identifier: oai:manglar.uninorte.edu.co:10584/12196
• Acceso en línea: http://hdl.handle.net/10584/12196
• Palabra clave: Modelos Predictivos
• License: Universidad del Norte