PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 7, 2026Alexandria Engineering Journal1 citationsOpen Access

GDBERT-score: Semantic Graph-Enhanced DeBERTa for automated essay scoring in higher education

View Full Paper
TMTiantian MiShandong Women’s UniversityTYTianjiao YuQingdao University

Key Points

  • The research aims to develop GDBERT-Score, a novel architecture for automated essay scoring that integrates various text modeling approaches.
  • Hybrid architecture combining DeBERTa and graph convolutional networks
  • Essays represented as semantic graphs with sentence embeddings as nodes
  • Evaluated on Kaggle Automated Essay Scoring 2.0 using six-fold cross-validation
  • GDBERT-Score achieved QWK of 0.7777 ± 0.0017, surpassing baseline models
  • Node feature quality more significant than graph topology in scoring effectiveness

Abstract

Automated essay scoring (AES) systems typically rely on either sequential text models for contextual semantics or graph-based approaches for discourse structure, but rarely integrate both effectively. We present GDBERT-Score, a hybrid architecture that combines DeBERTa’s disentangled attention with graph convolutional networks (GCN) for holistic essay scoring. Each essay is represented as a sentence-level semantic graph, where nodes are DeBERTa mean-pooled sentence embeddings and edges connect sentence pairs whose cosine similarity exceeds a learned threshold ( τ = 0 . 4 ). GCN-derived structural embeddings are concatenated with DeBERTa’s document-level contextual embedding and fused through a progressive reduction network ( 896 → 512 → 256 → 128 ) to produce a final holistic score. Evaluated on the Kaggle Automated Essay Scoring 2.0 benchmark via six-fold cross-validation, GDBERT-Score achieves QWK = 0.7777 ± 0.0017 (mean ± std across 5 random seeds), significantly outperforming both the DeBERTa-only baseline ( p < 0 . 0001 ) and the TF–IDF graph variant ( p = 0 . 0010 ). Ablation experiments reveal that node feature quality, rather than graph topology, is the primary determinant of GCN effectiveness in automated essay scoring: replacing surface-level node features with DeBERTa sentence embeddings yields significant improvements over both the DeBERTa-only baseline ( Δ QWK = +0.0294, p < 0 . 0001 ) and the graph-augmented variant ( Δ QWK = +0.0239, p = 0 . 0010 ), confirmed across five random seeds (QWK = 0.7777 ± 0.0017).

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Mi et al. (2026) studied this question.

synapsesocial.com/papers/69fbe2f2164b5133a91a23b9https://doi.org/10.1016/j.aej.2026.04.019
Ask AI
Helpful
Bookmark
Share
View Full Paper