Using Machine Learning Techniques to Earlier Predict Student's Performance

Abstract

Education field is rich of data, and machine learning used in this field are increasing lately. Based on their first semester result, using machine learning techniques, the student's final year result (GPA) can be predicted. The data used in this experiment are from the computer science subjects, 6 subjects, 1 laboratories results and the GPA on their graduation year. The techniques used in this experiment are Generalized Linear Model, Deep Learning and Decision Tree. From this result, what are the important factors that impact on the result can be extracted to help the students prepared themselves earlier.

How Machine Learning (ML) is Transforming Higher Education: A Systematic Literature Review

“…For instance, studies like (Rista and Mukli, 2022) employed a ML approach to predict, analyze and evaluate the potential causes of student absenteeism. Additionally, several studies used data driven approaches to predict academic results such as exam scores, academic retention, degree completion and GPA (Tanuar et al 2019;Musso et al 2020). For instance, (Musso et al, 2020) using multilayer perceptron artificial neural network models with a backpropagation algorithm, classified levels of grade point average, academic retention, and degree completion outcomes in a sample of 655 college students.…”

Deep Learning para Previsão do Desempenho do Estudante: Um Mapeamento Sistemático da Literatura

“…Com relação a interpretação dos resultados do DL, a maioria dos trabalhos não comentou nada a respeito. Tanuar et al (2018) destacam que Decision Tree aindaé o modelo mais fácil para apresentar os resultados ao usuário. Botelho et al (2019) e Lee e Yeung (2019) relatam que apesar dos resultados serem menos interpretáveis e do aumento da complexidade, o uso de DL compensa pelas vantagens mencionadas e os resultados mais precisos.…”