Predicting and Reducing Dropout in Virtual Learning using Machine Learning Techniques: A Systematic Review

Tamada, Mariela Mizota; Netto, José Francisco de Magalhães; Lima, Dhanielly P. R. de

doi:10.1109/fie43999.2019.9028545

Cited by 21 publications

(15 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Third, the mechanisms underlying the models or algorithms should be set. Data mining and machine learning have been widely used in recent years to build such prediction models (Agrusti et al, 2019;Manrique et al, 2019;Tamada et al, 2019). While data mining generally refers to merely the identification of hidden patterns in large datasets, machine learning-based algorithms are unique in the sense that they can improve automatically via experience.…”

Section: Predicting Students At-riskmentioning

confidence: 99%

“…While data mining generally refers to merely the identification of hidden patterns in large datasets, machine learning-based algorithms are unique in the sense that they can improve automatically via experience. The latter achieve very high levels of accuracy while predicting dropout; a recent literature review had identified 19 machine learning-based models, all of which achieved an accuracy of 91% or higher (Tamada et al, 2019).…”

Section: Predicting Students At-riskmentioning

confidence: 99%

See 1 more Smart Citation

Insights of Instructors and Advisors into an Early Prediction Model for Non-Thriving Students

Hershkovitz

Ambrose

2022

Learning Analytics

View full text Add to dashboard Cite

In this qualitative study (N=6), we explored insights of first-year students’ instructors and advisors into an early identification system aimed at detecting non-thriving students in the context of an all-campus first-year orientation course for undergraduates. Following the development of that prediction model in a bottom-up manner, using a plethora of available data, we focus on how its end-users could help us understand the underlying mechanisms that drive the identification of non-thriving students. As findings suggest, participants were appreciative overall of the prediction and its timing and came up with various behaviours that could explain non-thriving, mostly motivation and engagement. They suggested additional data that could predict non-thriving, including background information, academic engagement, and learning habits.

show abstract

Section: Predicting Students At-riskmentioning

confidence: 99%

Section: Predicting Students At-riskmentioning

confidence: 99%

Insights of Instructors and Advisors into an Early Prediction Model for Non-Thriving Students

Hershkovitz

Ambrose

2022

Learning Analytics

View full text Add to dashboard Cite

show abstract

“…Quanto aos fatores que afetam o desempenho e a aprendizagem, o estudo relata que existem diversos fatores, tais como: percepção dos alunos em relação aos conteúdos abordados, história pessoal do aluno, auto eficácia ou metacognição, frequênciaàs aulas, idade dos alunos, status socioeconômico, dentre outros. Tamada, Netto and Lima (2019) realizaram uma revisão sistemática dos trabalhos que utilizam aprendizado de máquina para reduzir altas taxas de evasão em ambientes virtuais de aprendizagem. A revisão relata as técnicas de aprendizado de máquina utilizadas e identifica soluções propostas para reduzir a evasão no ensinoà distância.…”

Section: Trabalhos Relacionadosunclassified

“…Portanto, pode-se assumir que até o momento não há um reconhecimento sistemático sobre elementos indicadores de evasão que se fazem presentes nos AVAs. O estudo de Tamada, Netto and Lima (2019) concentrou-se em tec-1 Uma abordagem indutiva que cria modelos para descobrir automaticamente informações ocultas presentes nos dados dos alunos que podem ser utilizadas na melhoria da aprendizagem (ROMERO; VENTURA; GARCÍA, 2008).…”

Section: Trabalhos Relacionadosunclassified

Indicadores de Evasão em Ambientes Virtuais de Aprendizagem no contexto da Educação a Distância: Um Mapeamento Sistemático

Chicon

Paschoal²,

Roos‐Frantz

2021

RENOTE

View full text Add to dashboard Cite

Um dos desafios das instituições de ensino é reduzir os altos índices de evasão em cursos na modalidade a distância. Este artigo tem como objetivo identificar na literatura os trabalhos de pesquisa sobre indicadores de evasão no contexto da educação a distância e a relação destes indicadores com recursos e dados de ambientes virtuais de aprendizagem. Um mapeamento sistemático foi conduzido, no qual foram examinados 22 estudos primários. Os resultados obtidos revelam os principais indicadores de evasão no contexto da educação a distância, são eles: indicador comportamental, desempenho e demográfico. Os dados coletados pelos ambientes virtuais de aprendizagem, utilizados pelos indicadores, provêm de recursos como envio de tarefas, participação em fórum, interações em chat, respostas e interações em questionários, dentre outros.

show abstract

“…Machine learning has the advantage of constructing models using both categorical and numerical predictions by evaluating linear and nonlinear relationships between variables [25]. Recently, most researchers have used machine learning techniques, as it is highly efficient in prediction models in distance learning [26]. In Jebaseeli and Kirubakaran [27], the authors proposed a neural network-based algorithm is designed to improve the feed-forward network algorithm (IOPNW FFNN) by adding weights from the input layer to the output layer for the M-learning classification.…”

Section: Introductionmentioning

confidence: 99%

Prediction of student satisfaction on mobile-learning by using fast learning network

Sultan

Abdulateef

Shtyat

2022

IJEECS

View full text Add to dashboard Cite

The rapid advancement of mobile technologies over the past decade has had a significant impact on the appearance of M-learning applications. The research proposes the fast learning network model to investigate and identify the factors that affect student satisfaction in M-learning for the University of Tikrit students. The research model is conducted utilizing a questionnaire of 300 participating students based on variables. This research showed that the proposed model's perfor mance was superior to artificial neural network, <br /> k-nearest neighbors, and multilayer perceptron algorithms. The accuracy and specificity of predicting the student satisfaction coefficients in M-learning were 91.6% and 92.85%, respectively. The proposed findings demonstrate that diversity in the evaluation, teacher attitude and response, and quality of technology are key operators of student satisfaction.

show abstract

Predicting and Reducing Dropout in Virtual Learning using Machine Learning Techniques: A Systematic Review

Cited by 21 publications

References 16 publications

Insights of Instructors and Advisors into an Early Prediction Model for Non-Thriving Students

Insights of Instructors and Advisors into an Early Prediction Model for Non-Thriving Students

Indicadores de Evasão em Ambientes Virtuais de Aprendizagem no contexto da Educação a Distância: Um Mapeamento Sistemático

Prediction of student satisfaction on mobile-learning by using fast learning network

Contact Info

Product

Resources

About