Caracterización y segmentación de público objetivo para scouting universitario a partir de minería de datos

Since its creation, the Scouting and Promotion office from Los Andes University has focused its activities towards the improvement on the attention to applicants and to increase registered people for the programs offered by the university. Despite the implementation of multiple changes and improvements in terms of contact, care and information availability for applicants, actually, the amount of applicants registered is not increasing any more. Those improvements and other activities executed without success suggested an evaluation of the strategy used to prioritize, in which the office executes exclusively its activities for schools and education institutions with an A+ classification in the SABER 11 test. This strategy left most institutions without attention based on their level and it was designed based assuming that the students from well-classified schools have a higher probability to be accepted and subsequently enroll to the university. Also, the late results questioned the knowledge of the influencing variables to the register and enrollment process. This document exposes the influencing variable identification process and a new list of schools to focus on based the identification made. To achieve this, a data mining model was built aiming to solve the actual registered applicants problem and to improve the conversion rates between applicants to enrolled students.

• Autores: Morales Guerrero, Felipe
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2015
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: spa
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/51668
• Acceso en línea: http://hdl.handle.net/1992/51668
• Palabra clave: Universidad de los Andes (Colombia); Minería de datos; Aprendizaje automático (Inteligencia artificial); Acceso a la educación; Ingeniería
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/