Three word embedding methods and three classifiers were introduced to predict item quality for accessible math assessments¶

Claim · All claims
Evidence · 1 study · associational r? · q2
Currency · newest study 2022 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 2025: tool and model accuracy changes quickly

Subclaims¶

q2 i? The researchers introduced three word embedding methods and three classifiers to predict item quality for accessible assessments. → Kang Xue 2022

Evidence¶

Kang Xue 2022¶

Kang Xue, Elizabeth Barker. (2022). Using Artificial Intelligence (AI) to improve math accessibility for students with visual impairments. https://www.nwea.org/resources/using-artificial-intelligence-ai-to-improve-math-accessibility-for-students-with-visual-impairments/

q2 · i? · associational · r?

The brief reports that "The researchers introduced three word embedding methods and three classifiers to predict item quality for accessible assessments." No performance statistics for the prediction models are printed in the brief.

"The researchers introduced three word embedding methods and three classifiers to predict item quality for accessible assessments."

Discussion¶

Related Claims¶

• The item-quality and prediction work is part of a larger project aiming to make online math more accessible to students with visual impairments — a broader claim this one bears on
• Math assessment items can be classified as high-quality or low-quality for students with visual impairments based on whether they measure achievement equally well for students with and without VI — related
• A random forest classified flagged versus non-flagged word problems with the best accuracy of five tested models (AUC = 0.75) — a narrower finding that bears on this claim
• AI pilots found models not yet capable of high-quality math output generation, including multiple choice problems, coherent word problems, and math visuals — related