Answers are grounded in this paper and cite their sources.
Methodological approach estimates knowledge similarity in automated learning systems, highlighting algorithmic advances.
This article examines methods for estimating the degree of similarity (distance) between student and reference responses as a primary step in assessing knowledge in automated learning systems. The work represents student and reference responses by string images (i.e., a sequence of numeric, alphanumeric, or symbolic primitives). Distance estimation is considered as the required number of transformations (i.e., insertions, deletions, and substitutions of primitives) required for obtaining the reference response string from the student’s response string. The distance estimation algorithm takes into account the non-equivalence of primitives and their various representations, and presents an algorithm that allows for responses in natural language.
No takes yet. Share an insight, caveat, or question.
A 2026 study studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: