Identificação dos Fatores que Influenciam a Evasão em Cursos de Graduação Através de Sistemas Baseados em Mineração de Dados: Uma Abordagem Quantitativa

Manhães, Laci Mary Barbosa; Cruz, Sérgio M. Serra Da; Costa, Raimundo José Macário; Zavaleta, Jorge; Zimbrão, Geraldo

doi:10.5753/sbsi.2012.14413

Cited by 6 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Em Manhães et al [6] foi utilizado o modelo Naive Bayes para visualizar os fatores que distinguem os alunos que tiveram sucesso ou fracasso em seus cursos. Nesse contexto, foram analisados dados referentes ao período de 2003 a 2014 de 155 cursos de graduação de 28 unidades da Universidade Federal do Rio de Janeiro (UFRJ).…”

Section: Trabalhos Relacionadosunclassified

Desenvolvimento de um Score para análise de risco de evasão de estudantes do Ensino Superior baseado em Aprendizado de Máquina

Crusoé da Cruz,

Correa Juliano,

Cristinne Correa Souza

et al. 2022

Anais Do XIII Computer on the Beach - COTB'22

View full text Add to dashboard Cite

Dropping out of Higher Education contributes to great social, eco- nomic and academic loss. Among the main reasons for dropping out are the student’s difficulty in following the content, the struc- ture proposed by the course and the lack of financial resources. In recent years, several studies have emerged to try to identify groups of students at risk of dropping out, either by identifying the factors that can contribute to dropout, or by creating classifiers based on Machine Learning. However, researches focus essentially on categorical indicators, that is, with binary results, which denote that the student is or is not in the risk group. This type of anal- ysis is important, however, it does not show the variation in the student’s performance during their academic life, in addition to not offering a score within a performance score. Differently, this project use Machine Learning techniques in the creation of a Score, in order to provide a thermometer to analyze how close the student is or not to the dropout group. Preliminary results are promising, because when using KNN to create the Score, it was possible to develop a Score with the best result of hyperparameters found in the experiments.

show abstract

Section: Trabalhos Relacionadosunclassified

Desenvolvimento de um Score para análise de risco de evasão de estudantes do Ensino Superior baseado em Aprendizado de Máquina

Crusoé da Cruz,

Correa Juliano,

Cristinne Correa Souza

et al. 2022

Anais Do XIII Computer on the Beach - COTB'22

View full text Add to dashboard Cite

show abstract

“…Na etapa de classificação, os estudos citados indicam a utilização de aprendizagem supervisionada. Nesse processo os dados minerados (amostras) são aplicados no treinamento dos algoritmos de classificação, que resulta em um modelo para identificar novos dados cuja classe é desconhecida (MANHAES et al, 2012). Em seguida, é possível verificar a eficácia do modelo através de um conjunto de dados de testes, calculando a medida que representa a proximidade do valor obtido pelo modelo e o valor verdadeiro (acurácia) (MANHAES et al, 2012).…”

Section: Técnicas Para Identificação Dos Dados De Evasãounclassified

“…Já Manhaes et al (2012) aplicou 6 algoritmos de classificação para validar qual deles se comporta melhor no conjunto de dados utilizado no seu estudo, tendo como melhor rendimento o algoritmo de Árvore de Decisão, com resultado de 83.06% de acurácia. Segundo o autor, embora o Naive Bayes tenha obtido acurácia menor (80.71%) do que os outros classificadores, ele atendia ao objetivo do seu trabalho.…”

Section: Técnicas Para Identificação Dos Dados De Evasãounclassified

Evasão Escolar Universitária e Estratégias de Intervenções para Retenção do Estudante: Um Estudo de Caso na Universidade Federal de São Carlos

Santos¹

View full text Add to dashboard Cite

School dropout is a concern for educational institutions, especially higher education, since it directly impacts management and academic results of institutions, as well as being directly related to social problems. The literature points out that analyzing this phenomenon is a positive factor for developing programs to combat dropout, in addition to planning interventional actions and academic monitoring. Studies shows positive results in the use of Machine Learning techniques for the early identification of students before they dropout, based on the exploration of data from academic systems. Although there are studies with Machine Learning, there were no records of works focused on developing intervention strategies supported by the visualization of academic data. This work aimed to understand the overview of a Brazilian public university, UFSCar, using the exploration and classification of academic data through Machine Learning techniques. The analysis of the data allowed us to obtain an overview of the university's dropout data and made it possible to prepare digital reports with information and statistics to assist university managers, heads of centers and departments, course coordinators and teachers in decision making. In addition, the main stakeholders were interviewed to report their difficulties in knowing the statistics on dropout and to validate the premises initially raised in this work. The reports were evaluated by these stakeholders and resulted in positive perceptions of use. A second intervention was conducted with students in partnership with ProEstudo, a program composed of Psychology professionals from the university. The partnership allowed the improvement of a computational solution, ESPIM, capable of carrying out remote interventions, thus enabling the development of intervention models to support the monitoring of students to combat academic difficulties and dropout. The remote interventions are being used by ProEstudo professionals during the university's academic period. The results of this research point out that data exploration is fundamental to obtain reliable information and visualize the dropout scenario in the institution; in addition, it was possible to confirm that the interventions carried out provide sufficient means to assist managers in decision making and support professionals in monitoring the students, which may result in reduced dropout rates.

show abstract

Analyzing College Student Dropout Risk Prediction in Real Data Using Walk-Forward Validation

Santos,

Ponti,

Rodrigues

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Identificação dos Fatores que Influenciam a Evasão em Cursos de Graduação Através de Sistemas Baseados em Mineração de Dados: Uma Abordagem Quantitativa

Cited by 6 publications

References 5 publications

Desenvolvimento de um Score para análise de risco de evasão de estudantes do Ensino Superior baseado em Aprendizado de Máquina

Desenvolvimento de um Score para análise de risco de evasão de estudantes do Ensino Superior baseado em Aprendizado de Máquina

Evasão Escolar Universitária e Estratégias de Intervenções para Retenção do Estudante: Um Estudo de Caso na Universidade Federal de São Carlos

Analyzing College Student Dropout Risk Prediction in Real Data Using Walk-Forward Validation

Contact Info

Product

Resources

About