ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring

Essay writing tasks are crucial for assessing students’ writing skills, but manual evaluation can be time-consuming and prone to inconsistencies. Automated Essay Scoring (AES) offers a solution by automatically evaluating essays, reducing the need for human intervention. This paper presen...

Full description

Saved in:
Bibliographic Details
Main Authors: Hind Aljuaid, Areej Alhothali, Ohoud Alzamzami, Hussein Assalahi, Tahani Aldosemani
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10945775/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702076535275520
author Hind Aljuaid
Areej Alhothali
Ohoud Alzamzami
Hussein Assalahi
Tahani Aldosemani
author_facet Hind Aljuaid
Areej Alhothali
Ohoud Alzamzami
Hussein Assalahi
Tahani Aldosemani
author_sort Hind Aljuaid
collection DOAJ
description Essay writing tasks are crucial for assessing students’ writing skills, but manual evaluation can be time-consuming and prone to inconsistencies. Automated Essay Scoring (AES) offers a solution by automatically evaluating essays, reducing the need for human intervention. This paper presents a hybrid method, called Ensemble Transformer-Based Graph Neural Networks (ET-GNN), which integrates Transformer-based models with Graph Convolutional Networks (GCNs) for holistic AES. Three Transformer models, DistilBERT, RoBERTa, and DeBERTaV3, were individually fine-tuned to generate contextual embeddings for each essay. The GCNs process these embeddings, effectively capturing relevant semantic information and inter-essay similarities. Additionally, ensemble methods are used to combine the DistilBERT-GCN, RoBERTa-GCN, and DeBERTaV3-GCN models employing averaging for regression tasks, majority voting for classification tasks, and a weighted ensemble method for both types of tasks. The proposed ET-GNN method enhances the performance and robustness of AES systems, achieving Quadratic Weighted Kappa (QWK) scores of 0.835 and 0.841 on the ASAP and AES 2.0 datasets, respectively. These results outperform other state-of-the-art models based on Transformer or GCNs architectures for the AES task.
format Article
id doaj-art-d225be0b239341d08adf46e2a512c417
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d225be0b239341d08adf46e2a512c4172025-08-20T03:17:46ZengIEEEIEEE Access2169-35362025-01-0113587465875810.1109/ACCESS.2025.355635210945775ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay ScoringHind Aljuaid0https://orcid.org/0009-0000-6451-1238Areej Alhothali1https://orcid.org/0000-0001-9727-0178Ohoud Alzamzami2Hussein Assalahi3Tahani Aldosemani4https://orcid.org/0000-0002-2347-1564Department of Computer Science, King Abdulaziz University, Jeddah, Saudi ArabiaDepartment of Computer Science, King Abdulaziz University, Jeddah, Saudi ArabiaDepartment of Computer Science, King Abdulaziz University, Jeddah, Saudi ArabiaEnglish Language Institute, King Abdulaziz University, Jeddah, Saudi ArabiaCollege of Education, Prince Sattam bin Abdulaziz University, Al-Kharj, Saudi ArabiaEssay writing tasks are crucial for assessing students’ writing skills, but manual evaluation can be time-consuming and prone to inconsistencies. Automated Essay Scoring (AES) offers a solution by automatically evaluating essays, reducing the need for human intervention. This paper presents a hybrid method, called Ensemble Transformer-Based Graph Neural Networks (ET-GNN), which integrates Transformer-based models with Graph Convolutional Networks (GCNs) for holistic AES. Three Transformer models, DistilBERT, RoBERTa, and DeBERTaV3, were individually fine-tuned to generate contextual embeddings for each essay. The GCNs process these embeddings, effectively capturing relevant semantic information and inter-essay similarities. Additionally, ensemble methods are used to combine the DistilBERT-GCN, RoBERTa-GCN, and DeBERTaV3-GCN models employing averaging for regression tasks, majority voting for classification tasks, and a weighted ensemble method for both types of tasks. The proposed ET-GNN method enhances the performance and robustness of AES systems, achieving Quadratic Weighted Kappa (QWK) scores of 0.835 and 0.841 on the ASAP and AES 2.0 datasets, respectively. These results outperform other state-of-the-art models based on Transformer or GCNs architectures for the AES task.https://ieeexplore.ieee.org/document/10945775/Automated essay scoring (AES)transformer modelsgraph convolutional networks (GCNs)ensemble methodsnatural language processing (NLP)
spellingShingle Hind Aljuaid
Areej Alhothali
Ohoud Alzamzami
Hussein Assalahi
Tahani Aldosemani
ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring
IEEE Access
Automated essay scoring (AES)
transformer models
graph convolutional networks (GCNs)
ensemble methods
natural language processing (NLP)
title ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring
title_full ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring
title_fullStr ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring
title_full_unstemmed ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring
title_short ET-GNN: Ensemble Transformer-Based Graph Neural Networks for Holistic Automated Essay Scoring
title_sort et gnn ensemble transformer based graph neural networks for holistic automated essay scoring
topic Automated essay scoring (AES)
transformer models
graph convolutional networks (GCNs)
ensemble methods
natural language processing (NLP)
url https://ieeexplore.ieee.org/document/10945775/
work_keys_str_mv AT hindaljuaid etgnnensembletransformerbasedgraphneuralnetworksforholisticautomatedessayscoring
AT areejalhothali etgnnensembletransformerbasedgraphneuralnetworksforholisticautomatedessayscoring
AT ohoudalzamzami etgnnensembletransformerbasedgraphneuralnetworksforholisticautomatedessayscoring
AT husseinassalahi etgnnensembletransformerbasedgraphneuralnetworksforholisticautomatedessayscoring
AT tahanialdosemani etgnnensembletransformerbasedgraphneuralnetworksforholisticautomatedessayscoring