Imbalanced Learning Techniques for Improving the Performance of Statistical Models in Automated Essay Scoring

Filho, Aluízio Haendchen; Concatto, Fernando; Nau, Jonathan; Prado, Hércules Antônio do; Imhof, Diego Oscar; Ferneda, Edílson

doi:10.1016/j.procs.2019.09.235

Cited by 6 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Some researchers applied RUS, NearMiss, SMOTEENN, and SMOTETL [67] to overcome this problem to improve the imbalanced classification. A study in [52] proposed four algorithms based on oversampling and undersampling techniques to minimize the effects of imbalanced problems in predicting essay grading. The distribution of each class is based on a stratified sampling approach so that the training dataset is maintained in the same ratios.…”

Section: Data-level Oversampling Smotementioning

confidence: 99%

Imbalanced Classification Methods for Student Grade Prediction: A Systematic Literature Review

et al. 2023

View full text Add to dashboard Cite

Student success is essential for improving the higher education system student outcome. One way to measure student success is by predicting students' performance based on their prior academic grades. Concerning the significance of this area, various predictive models are widely developed and applied to help the institution identify students at risk of failure. However, building a high-accuracy of the predictive model is a challenge due to the imbalanced nature of the dataset, which caused to bias results. Therefore, this study aims to review the existing research article by providing a state-of-the-art approach for handling imbalanced classification in higher education, including the best practices of dataset characteristics, methods, and comparative analysis of the proposed algorithms, focusing on student grade prediction context problems. The study also presents the most common balancing methods published from 2015 to 2021 and highlights their impact on resolving imbalanced classification in three approaches: data-level, algorithm-level, and hybridlevel. The survey results reveal that the data-level approach using SMOTE oversampling is broadly applied in determining imbalanced problems for student grade prediction. However, the application of hybrid and feature selection methods supporting the generalization of the predictive model to boost student grade prediction performance is generally lacking. Other than that, some of the strengths and weaknesses of the proposed methods are discussed and summarized for the direction of future research. The outcomes of this review will guide the professionals, practitioners, and academic researchers in dealing with imbalanced classification, mainly in the higher education field.

show abstract

Section: Data-level Oversampling Smotementioning

confidence: 99%

Imbalanced Classification Methods for Student Grade Prediction: A Systematic Literature Review

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Essas pesquisas, em geral, buscam estimar um valor numérico (nota) representando a qualidade global da redac ¸ão com base em diferentes critérios. No contexto das provas no estilo do ENEM, existem trabalhos que buscam estimar a nota geral obtida pela redac ¸ão [Filho et al 2019, Costa et al 2020, Filho. et al 2021] ou aqueles que focam em competências específicas [Lima et al 2018, Passero et al 2019.…”

Section: Trabalhos Relacionadosunclassified

“…Para tentar mitigar esse problema foram utilizados os seguintes métodos de balanceamento disponíveis na biblioteca Imbalanced-learn 7 : Sobreamostragem aleatória, do inglês Random Oversampling, e a Técnica de sobreamostragem minoritária sintética, do inglês Synthetic Minority Oversampling Technique (SMOTE). Esses métodos foram escolhidos por serem comumente usados na literatura em problemas de aprendizado supervisionado em bases de dados desbalanceadas [Kaur et al 2019, Filho et al 2019. Essas técnicas foram utilizadas somente na etapa de treinamento durante o processo de validac ¸ão cruzada, resultando em um conjunto de treino balanceado contendo 1.600 redac ¸ões para cada um dos grupos de notas (0-200).…”

Section: Selec ¸ãO E Avaliac ¸ãO Dos Modelos De Regressãounclassified

Estimando Coesão Textual em Redações no Contexto do ENEM Utilizando Modelos de Aprendizado de Máquina

Oliveira

Miranda²,

Isotani³

et al. 2022

Anais Do XXXIII Simpósio Brasileiro De Informática Na Educação (SBIE 2022)

View full text Add to dashboard Cite

Coesão textual é uma propriedade fundamental da escrita formal, pois tem relação com a conexão harmoniosa dos elementos de um texto. Apesar de diversos trabalhos analisarem automaticamente a coesão textual em redações, ainda são escassos trabalhos para o português. Este trabalho investiga modelos de regressão para estimar a coesão textual de redações escritas em português no contexto do ENEM, adotando um conjunto de 151 características identificadas na literatura. Experimentos usando a base de dados do Essay-BR, composta por 4.570 redações no estilo do ENEM, demonstram que o modelo de Extremely Randomized Trees apresentou os melhores resultados com uma correlação de Pearson (53,08%) moderada com as notas relacionadas à coesão.

show abstract

“…However, this study has used an imbalance dataset where five scores are used as 0, 50, 100, 150, and 200. The classes of the students' answers were imbalanced therefore, latterly the authors have proposed an improvement in their work of [14]. Using some statistical algorithms, the authors have managed to improve the accuracy of assessment by obtaining 75% of correlation.…”

Section: A Supervised Aesmentioning

confidence: 99%

“…Usually, this type of feature analysis is exploited by the unsupervised technique through a ranking procedure that gives score for each criterion [22], or it could be exploited within a set of rules ([2]; [5]; [12]). On the other hand, this feature analysis can be utilized by a supervised technique through the ranking procedure where the numeric ranks would be fed to a regressor ( [14]; [15]). www.ijacsa.thesai.org…”

Section: A Structurementioning

confidence: 99%

Automatic Essay Scoring: A Review on the Feature Analysis Techniques

Chassab¹,

Zakaria²,

Tiun³

2021

IJACSA

View full text Add to dashboard Cite

Imbalanced Learning Techniques for Improving the Performance of Statistical Models in Automated Essay Scoring

Cited by 6 publications

References 7 publications

Imbalanced Classification Methods for Student Grade Prediction: A Systematic Literature Review

Imbalanced Classification Methods for Student Grade Prediction: A Systematic Literature Review

Estimando Coesão Textual em Redações no Contexto do ENEM Utilizando Modelos de Aprendizado de Máquina

Automatic Essay Scoring: A Review on the Feature Analysis Techniques

Contact Info

Product

Resources

About