Cap Cloze questions per key sentence at the three highest-probability candidate words to reduce manual evaluation burden¶

Design · All designs
Evidence · no claims cited

Description¶

The article limits blank selection to the top-ranked words from the Naive Bayes probability list, because generating every candidate would make human evaluation prohibitive: "the number of Cloze questions generated from each key sentence is limited by choosing only the first three words from the Word_prd list." This keeps expert evaluation of usefulness tractable while retaining the highest-probability blanks.

Design Implications¶

Context¶

Requirements¶

A sorted Word_prd list of words with computed blank-substitution probabilities

Constraints¶

The article presents this cap as an evaluation-burden trade-off; it does not report testing whether three blanks per sentence is optimal

Target Learners¶

students taking MCQ assessments in technical courses

Learning Goals¶

practical scale-up of automatically generated Cloze items for drills and practice

Claims¶

Mbt Three Stage Cloze Pipeline [+M]

Examples¶

-

Key Sources¶

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

Page metadata

Field	Value
`type`	`design`
`id`	`cap-cloze-blanks-top-three-per-sentence`
`title`	`Cap Cloze questions per key sentence at the three highest-probability candidate words to reduce manual evaluation burden`
`description`	`The article limits blank selection to the top-ranked words from the Naive Bayes probability list, because generating every candidate would make human evaluation prohibitive: "the number of Cloze questions generated fr...`
`status`	`review`
`generated`	by `process:settle-candidates` on `2026-10-28`
`trust`	unverified — no one has confirmed this page's content
`sources`	1 — `praveen-kumar-2024`

Frontmatter is the page's machine-readable record — see the Schema & Guide.