Citas bibligráficas
Montilla, H., (2022). Algoritmos de aprendizaje automático supervisado en la predicción del rendimiento académico [Universidad Nacional de San Martín. Fondo Editorial]. http://hdl.handle.net/11458/4667
Montilla, H., Algoritmos de aprendizaje automático supervisado en la predicción del rendimiento académico []. PE: Universidad Nacional de San Martín. Fondo Editorial; 2022. http://hdl.handle.net/11458/4667
@mastersthesis{renati/1055483,
title = "Algoritmos de aprendizaje automático supervisado en la predicción del rendimiento académico",
author = "Montilla Garcia, Henrry",
publisher = "Universidad Nacional de San Martín. Fondo Editorial",
year = "2022"
}
The general objective was to analyze the efficiency of supervised machine learning algorithms in the prediction of academic performance in mathematics of students in the fifth grade of secondary school at I. E. "Santa Rosa". The specific objectives were: to predict academic performance by 2021 with the most efficient supervised machine learning algorithm; to determine the relationship between the 2010 to 2020 scores and the prediction of academic performance by 2021; and to forecast academic performance in mathematics from 2021 to 2027. The research was basic, descriptive, composite, predictive, non-experimental. The census sample was made up of 2933 students. The general hypothesis stated that there are supervised machine learning algorithms that predict the academic performance in mathematics of fifth grade students, with an efficiency greater than or equal to 95%. The research concluded that there are supervised machine learning algorithms that predict academic performance with efficiencies greater than or equal to 95%. In addition, academic performance prediction for 2021 in mathematics is reliable with 100% accuracy, using the K nearest neighbors supervised machine learning algorithm as the most efficient; no significant relationship exists between the 2010 to 2020 scores and the prediction of academic performance for 2021 with the K nearest neighbors algorithm; in addition, the forecast of academic performance in mathematics from 2021 to 2027 has a positive and incremental trend.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons