DeBERTa achieved the highest overall accuracy (86.36%) among the evaluated NLP models for classifying learner-generated questions, yet struggled with contextual nuances in complex queries¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? DeBERTa achieved the highest overall accuracy (86.36%) classifying learner-generated questions but struggled with capturing contextual nuances inherent in complex queries. → Lee 2026
Evidence¶
Lee 2026¶
Lee, H., Atif, A., & Kang, K. (2026). Analysing AI utilisation in education through learner question types: A constructivist approach. Australasian Journal of Educational Technology, 42(1), 79–96. https://doi.org/10.14742/ajet.10657
q2 · i? · design · r2
Model evaluation study comparing several NLP models (BERT baseline and fine-tuned, DeBERTa, RoBERTa) on classifying learner-generated questions. The article reports that "DeBERTa achieved the highest overall accuracy (86.36%)" while noting its difficulty with contextual nuance in complex queries.
"Notably, DeBERTa achieved the highest overall accuracy (86.36%) yet struggled with capturing contextual nuances inherent in complex queries."
Discussion¶
Related Claims¶
- NLP models classify factual (knowledge transmitter) learner questions well but face challenges distinguishing higher-order facilitator and co-learner questions — related
- Question and observed-coverage features retrospectively predict high-divergence questions with AUC 0.774, but the model is not validated for new-query routing — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
deberta-highest-accuracy-86-36-learner-question-classification |
title |
DeBERTa achieved the highest overall accuracy (86.36%) among the evaluated NLP models for classifying learner-generated questions, yet struggled with contextual nuances in complex queries |
description |
DeBERTa achieved the highest overall accuracy (86.36%) among the evaluated NLP models for classifying learner-generated questions, yet struggled with contextual nuances in complex queries |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
generated |
by process:wiki-ingest on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — lee-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.