Optimizing Automated Essay Scoring: A Comparative Study of Machine Learning Approaches with a Focus on Ensemble Methods
Bibliographic record
Abstract
This study examines the optimization of Automated Essay Scoring (AES) systems for English language writing using advanced machine learning techniques, focusing on ensemble methods to enhance accuracy, consistency, and interpretability. An English written corpus includes a total of 17,793 English essays: 12,976 from the Automated Student Assessment Prize (ASAP) dataset and 4,817 from the Khon Kaen University Academic English Language Test (KKU-AELT). Linguistic features and semantic content critical to English writing proficiency were assessed using BERT, XGBoost, and Neural Networks models. Combining these models with Ridge Regression, the ensemble approach substantially reduced Root Mean Squared Error (RMSE) while balancing Cohen's Kappa and Quadratic Weighted Kappa scores, highlighting interpretive alignment challenges. The SHAP values were employed for feature importance analysis, and Bayesian optimization was applied for hyperparameter tuning, enhancing model transparency. The findings highlight the potential of ensemble AES to evaluate diverse aspects of English such as argumentation, coherence, and vocabulary complexity—applicable to various domains, from applied linguistics to literature and translation studies. The research offers scalable solutions for teaching and assessment, aligning AES systems with the pedagogical goals of supporting skill acquisition and providing actionable feedback. The study concludes that advanced AES models can serve as valuable complementary tools in language assessment, assisting teachers by providing consistent, detailed insights that foster English writing proficiency and skills development across diverse educational contexts.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.027 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".