Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/147542
Title: Segmentación del estudiantado universitario: el caso de la Udima
Author: Bonilla Garzón, Alejandra
Tutor: Subirats, Laia  
Others: Prados Carrasco, Ferran  
Abstract: This final project ascertains the different typologies of Udima students based on their performance, opinions, socio-demographic and behavioural characteristics in Moodle in order to customize their follow-up. The research population is made up of Udima university students (Bachelor's/Master's) enrolled in a course from the beginning (2008-09) to the final year (2021-22). The profile of 30.875 students who finish/leave their studies is identified using the optimal unsupervised learning technique: k-means group algorithm, Euclidean distance and 4 clusters obtaining: 25.09% of the group as dropout; 59.59% as common graduate; 8.76% as motivated: specialist graduate, with a sense of belonging to the university and high satisfaction and 6.55% as dissatisfied: graduate with below-average satisfaction. This classification is used to predict, by means of monitored learning techniques, the classification of 7.989 students in progress with a reduced dataset (variables related to graduation/dropout are eliminated so as not to condition). Using 10-Fold-Cross-Validation (and according to maximum accuracy), the Random Forest classification algorithm is applied. After training, an accuracy of 86.45% is obtained. The model is applied and the distribution is: dropout, 59.36%; common graduate, 30.85%; motivated, 1.11% and dissatisfied, 8.69%. After evaluating the results in relation to the problem raised and the results of the literature, the model is considered valid. The difference between this model and those used in Udima is the integration of information from satisfaction and job placement. The results obtained are specific to Udima, but the methodology used can be adapted to any advanced e-learning institution.
Keywords: segmentation
higher education
clustering
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jan-2023
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
abonillagaTFM0123presentacion.pdfPresentación del TFM3,88 MBAdobe PDFThumbnail
View/Open
abonillagaTFM0123memoria.pdfMemoria del TFM662,44 kBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons