RBASS: Rubric Based Automated Short Answer Scoring System

Ramesh Dadi, Suresh Kumar Sanampudi

2025

Abstract

Automated short-answer scoring is a crucial tool in the educational system, enabling the quick and efficient assessment of students' responses. This process helps alleviate the challenges associated with manual evaluation while enhancing the reliability and consistency of assessments. However, it raises questions regarding whether it adequately considers the content and coherence of responses. Many researchers have tackled this issue and achieved promising results through natural language processing and deep learning advancements. Nevertheless, some researchers have focused solely on coherence evaluation, neglecting to test their models against adversarial responses, which limits the robustness of the model. This paper proposes a novel RBASS approach to evaluating answers based on coherence and content. Additional rubrics were incorporated into the existing dataset for content-based evaluation, resulting in optimal outcomes. The study comprehensively reviews and compares the system's performance using quantitative and qualitative metrics. Additionally, the model's performance is evaluated rigorously by training and testing it across multiple datasets and subjecting it to adversarial responses. The findings indicate that the model performs optimally both quantitatively and qualitatively, highlighting its effectiveness in assessing short-answer responses.

Paper Citation

in Harvard Style

Dadi R. and Sanampudi S. (2025). RBASS: Rubric Based Automated Short Answer Scoring System. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 284-291. DOI: 10.5220/0013590900004664

in Bibtex Style

@conference{incoft25,
author={Ramesh Dadi and Suresh Kumar Sanampudi},
title={RBASS: Rubric Based Automated Short Answer Scoring System},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT},
year={2025},
pages={284-291},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013590900004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT
TI - RBASS: Rubric Based Automated Short Answer Scoring System
SN - 978-989-758-763-4
AU - Dadi R.
AU - Sanampudi S.
PY - 2025
SP - 284
EP - 291
DO - 10.5220/0013590900004664
PB - SciTePress