Detección de patrones académicos en curso de nivelación con deserción en la Universidad Técnica Estatal de Quevedo basado en los factores socioeconómicos
Currently, universities in Ecuador face great challenges due to student dropout, so it is urgent to take measures to identify the factors that affect this problem. To achieve the objectives, a descriptive and exploratory analysis was carried out on a set of data obtained from the Academic Management...
Furkejuvvon:
| Váldodahkki: | |
|---|---|
| Materiálatiipa: | masterThesis |
| Giella: | spa |
| Almmustuhtton: |
2024
|
| Fáttát: | |
| Liŋkkat: | https://repositorio.uteq.edu.ec/handle/43000/7819 |
| Fáddágilkorat: |
Lasit fáddágilkoriid
Eai fáddágilkorat, Lasit vuosttaš fáddágilkora!
|
Lasit fáddágilkoriid | Čoahkkáigeassu: | Currently, universities in Ecuador face great challenges due to student dropout, so it is urgent to take measures to identify the factors that affect this problem. To achieve the objectives, a descriptive and exploratory analysis was carried out on a set of data obtained from the Academic Management System department of the State Technical University of Quevedo. Through this project it is proposed to address the problem by implementing Data Mining algorithms with the aim of discovering patterns that cause student dropout. The knowledge discovery in database (KDD) methodology was used, which consists of selection, preprocessing, transformation, data mining and data evaluation phases. For the preparation and purification of the data set, RStudio and Weka are used to apply the J48, DecisionStump, RandomTree, RandomForest, HoeffdingTree, LMT and RepTree algorithms. To choose the optimal algorithm for the study, each of them will be quantitatively evaluated by precision. In the results obtained, it was highlighted that the Random Tree algorithm had better results in the analyses, followed by the Random Forest, this algorithm allowed us to identify patterns directly associated with student dropout. |
|---|