Prior AQG work shows ML yields grammatically correct Cloze questions while ontology approaches struggle with Cloze but produce meaningful Wh-type questions¶
Claim · All claims
Evidence · 1 study (2 entries) · designr2·q2
Currency · newest study 2024 · 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? Ontology-based approaches fail to generate grammatically correct Cloze questions compared to ML techniques because verbalizing ontology into Cloze questions yields grammatically incorrect questions. → Praveen Kumar 2024
q2 i? ML approaches generate reasonably good Cloze questions but fail to generate grammatically correct Wh-type questions compared to the ontology technique. → Praveen Kumar 2024 (2)
Evidence¶
Praveen Kumar 2024¶
Praveen Kumar, A., Nayak, A., Shenoy K., M., Chaitanya, & Ghosh, K. (2024). A Novel Framework for the Generation of Multiple Choice Question Stems Using Semantic and Machine-Learning Techniques. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-023-00333-6
q2 · i? · design · r2
The article's background synthesis, citing Kurdi et al. (2020) and Faizan & Lohmann (2018), reports that "verbalizing ontology into Cloze questions generates grammatically incorrect questions", motivating the assignment of Cloze generation to ML. This is the review's attribution, not a new experiment.
"The main reason is that verbalizing ontology into Cloze questions generates grammatically incorrect questions (Faizan & Lohmann, 2018)."
Praveen Kumar 2024 (2)¶
Praveen Kumar, A., Nayak, A., Shenoy K., M., Chaitanya, & Ghosh, K. (2024). A Novel Framework for the Generation of Multiple Choice Question Stems Using Semantic and Machine-Learning Techniques. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-023-00333-6
q2 · i? · design · r2
The article reports, citing the review by Ch and Saha (2018), that "ML generates reasonably good Cloze questions but fails to generate grammatically correct Wh-type questions", which motivates generating Wh-type stems from ontology semantics instead. Second-hand review attribution.
"ML generates reasonably good Cloze questions but fails to generate grammatically correct Wh-type questions compared to the ontology technique."
Discussion¶
Related Claims¶
- Existing AQG approaches concentrate on non-technical domains, and technical-domain Wh-type generation has been limited to Bloom's cognitive Level I — related
- Expert evaluation found the hybrid system's generated MCQ stems promising in usefulness and grammatical correctness — related
- EMCQG generated more than 3 million distinct case-based questions for four physician specialities from the EMMeT ontology — related
- Ontology-based generation produced exam-appropriate case-based MCQs, with 30% of a stratified sample rated appropriate by both expert raters and 50% by at least one — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
ml-cloze-ontology-wh-grammaticality-gap |
title |
Prior AQG work shows ML yields grammatically correct Cloze questions while ontology approaches struggle with Cloze but produce meaningful Wh-type questions |
description |
Prior AQG work shows ML yields grammatically correct Cloze questions while ontology approaches struggle with Cloze but produce meaningful Wh-type questions |
status |
review |
evidence_strength |
weak |
newest_evidence |
2024 |
generated |
by process:wiki-ingest on 2026-10-27 |
trust |
unverified — no one has confirmed this page's content |
sources |
2 — praveen-kumar-2024, praveen-kumar-2024-2 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.