Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
December 20, 2025International Journal of Information and Education TechnologyOpen Access

Harnessing Transformers for Enhancing Arabic Educational Assessment

View Full Paper
Ask AI
Bookmark
Share

Authors

ENEmad NabilIslamic University of MadinahMSMostafa SaeedNew York UniversityRRRana RedaMinistry of Communication and Information Technology

Discussion

Loading...

Member takes

Overview

This approach demonstrates a novel scoring method achieving over 92% correlation with student responses in Arabic assessments, suggesting advancements in grading accuracy.

Key Points

  • The research aims to develop a transformer-based model for grading Arabic short-answer questions.
  • Utilized the Cairo University Dataset focused on environmental science
  • Explored preprocessing strategies and multiple transformer models
  • Integrated models into a custom regression-based neural network
  • Achieved a Pearson correlation of 92.34% on the Cairo University Dataset
  • Outperformed existing benchmarks on the Arabic Short Answer Grading dataset with 80% Pearson correlation
  • Demonstrated potential for scalable and fair educational assessments

Cite This Study

Nabil et al. (2025) studied this question.

synapsesocial.com/papers/6945e9325151ab1219e4d67bhttps://doi.org/10.18178/ijiet.2025.15.12.2474
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Leveraging Transformers and LLMs for Automated Grading and Feedback Generation Using a Novel Dataset2026
  2. 2An Automatic Grading System for Arabic Language Short-Answer Questions Using Deep Learning2025
  3. 3Pre-Trained Transformer-Based Approach for Arabic Question Answering: A Comparative Study2025 · 1 citations
  4. 4Large Language Models for Arabic Automated Essay Scoring2026
  5. 5Auto-Grading Comprehension on Reference-Student Answer Pairs using the Siamese-based Transformer2024 · 4 citations