Prediction of academic results with the nntool application in Matlab using artificial neural networks
In 2018, Peru participated in the International Student Assessment (PISA), where disappointing results were evidenced in students of regular basic education. In this sense, the prediction of academic results was considered as an instrument for improving school performance. The objective of this rese...
Guardado en:
| Publicado en: | Apuntes Universitarios |
|---|---|
| Autores principales: | , , , , , |
| Formato: | Artículo de revista |
| Lenguaje: | español |
| Publicado: |
Universidad Peruana Unión
2021
|
| Materias: | |
| Acceso en línea: | Ver en el sistema de origen |
| Etiquetas: |
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | In 2018, Peru participated in the International Student Assessment (PISA), where disappointing results were evidenced in students of regular basic education. In this sense, the prediction of academic results was considered as an instrument for improving school performance. The objective of this research was to predict the annual average of second grade students of the Educational Institution N ° 16093 in the province of Jaén-Peru, through the design and implementation of an artificial neural network (ANN). To collect data on variables that influence the annual average of the student, a questionnaire with dichotomous responses was developed. In the validation and reliability, we used the expert judgment criterion and the Kuder-Richarson test respectively, the reliability coefficient obtained in a pilot test applied to 15 students was 0.8359. In the Matlab Scientific Software with the help of the nntool application, the RNA was designed consisting of three hidden layers and an output layer. The ANN during training, validation and testing, registered a weighted correlation coefficient of 0.967190, and a mean square error of 0.05011. The neural model implemented under the given conditions achieved an effectiveness of 88.670% and 98.522%. |
|---|---|
| ISSN: | 2304-0335 2225-7136 |