Contextual similarity scoring model for automated essay grading

Research Method · All research methods
Evidence · 2 claims (1 for, 1 against) · 2 studies (1 associational, 1 design), q1–q2 · 0 of 2 report an effect size · 2 claims rest on one study

Description

The article's scoring framework matches a student's response to an examiner-supplied model answer using semantic contextual search similarity, returning a similarity index of 0–1 that is multiplied by the item's allotted marks. Its matching phase "combines the similarity algorithm (answer matching), which are Generic Words (GW), Longest Common Subsequence (LCS), and Semantic Distance (SD)", weighted by each algorithm's strength. The model begins with tokenization or normalization of the two strings and considers all possible ways examinees could phrase responses, so it awards scores without manual training on many model answers.

Accounts

• Contextual similarity scoring model for automated essay grading: The article's scoring framework matches a student's response to an examiner-supplied model answer using semantic contextual search similarity, returning a similarity index of 0–1 that is multiplied by the item's allotted marks. Its matching phase "combines the similarity algorithm (answer matching), which are Generic Words (GW), Longest Common Subsequence (LCS), and Semantic Distance (SD)", weighted by each algorithm's strength. The model begins with tokenization or normalization of the two strings and considers all possible ways examinees could phrase responses, so it awards scores without manual training on many model answers. (Olaoye DD et al. (2026))

Claims

• AEEGS scores show high absolute agreement with 12 human expert raters' scores on mathematical Economics essay items (average-measures ICC = 0.863) [+M]
• NLI analysis reveals reference-based grading is limited: many semantically correct responses contradict reference answers [-W]

Related Research Methods

-

Key Sources

• Olaoye DD, Owolabi HO and Olaoye OT (2026). Automated software scoring of senior school certificate examination mathematical items in economics using a contextual similarity model. Front. Educ. 11:1669504. https://doi.org/10.3389/feduc.2026.1669504

Page metadata

Field	Value
type	research-method
id	contextual-similarity-scoring-model-for-automated-essay-grading
title	Contextual similarity scoring model for automated essay grading
description	The article's scoring framework matches a student's response to an examiner-supplied model answer using semantic contextual search similarity, returning a similarity index of 0–1 that is multiplied by the item's allotted marks.
status	review
generated	by process:settle-candidates on 2026-10-10
trust	unverified — no one has confirmed this page's content

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