Optimizing DeBERTa V3 for Accurate Automated Essay Evaluation in the Indonesian Educational Context

Muhammad Rahmat, Muhammad Niswar, Ady Wahyudi Paundu

2025

Abstract

Automated Essay Scoring (AES) has emerged as a practical solution to address the limitations of manual essay evaluation, particularly in educational contexts where scalability, consistency, and efficiency are critical. This study investigates the effectiveness of the DeBERTa v3 model for automatically scoring Indonesianlanguage essays. We fine-tuned the model using a dataset of student essays that were manually rated by human evaluators and optimized it with the Quadratic Weighted Kappa (QWK) metric to better align with ordinal scoring systems. Through 5-fold cross-validation, the model achieved an average QWK of 0.558, MAE of 0.748, and RMSE of 0.894. These results demonstrate that the model can generate consistent and reliable scores that closely approximate human judgment. The findings suggest that DeBERTa v3 holds significant potential for supporting automated assessment systems in Indonesian education and advancing natural language processing applications in low-resource languages.

Paper Citation

in Harvard Style

Rahmat M., Niswar M. and Paundu A. (2025). Optimizing DeBERTa V3 for Accurate Automated Essay Evaluation in the Indonesian Educational Context. In Proceedings of the 1st International Conference on Research and Innovations in Information and Engineering Technology - Volume 1: RITECH; ISBN 978-989-758-784-9, SciTePress, pages 217-223. DOI: 10.5220/0014272600004928

in Bibtex Style

@conference{ritech25,
author={Muhammad Rahmat and Muhammad Niswar and Ady Wahyudi Paundu},
title={Optimizing DeBERTa V3 for Accurate Automated Essay Evaluation in the Indonesian Educational Context},
booktitle={Proceedings of the 1st International Conference on Research and Innovations in Information and Engineering Technology - Volume 1: RITECH},
year={2025},
pages={217-223},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0014272600004928},
isbn={978-989-758-784-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Innovations in Information and Engineering Technology - Volume 1: RITECH
TI - Optimizing DeBERTa V3 for Accurate Automated Essay Evaluation in the Indonesian Educational Context
SN - 978-989-758-784-9
AU - Rahmat M.
AU - Niswar M.
AU - Paundu A.
PY - 2025
SP - 217
EP - 223
DO - 10.5220/0014272600004928
PB - SciTePress