Annotated student-utterance emotion dataset (224 turns) for text-based emotion recognition in math tutoring¶
Product or Programme · All products and programmes
Evidence · 2 claims (1 for, 1 mixed) · 1 study (1 design),q2· 0 of 1 report an effect size · 2 claims rest on one study
Description¶
Debanjana Kar et al.'s dataset contains 224 manually annotated MathDial-Bridge student turns with boredom, engagement, or neutral labels and polarity scores for developing emotion-recognition models in math tutoring dialogues.
Components¶
Claims¶
- MathBuddy's multimodal emotion aggregation matched participant-reviewed emotions at 60% overall accuracy, with sharply reduced recall for the Neutral class [~M] — accuracy of the MathBuddy emotion recognition this dataset was built to develop
- Adding student-emotion information to LLM tutor prompting produces more desirable pedagogical qualities across most evaluated dimensions, with larger models benefiting most [+M] — why recognising student emotion in text is worth doing
Related Products and Programmes¶
- Adapt tutor responses to detected learner affect — the use emotion labels like these support
Key Sources¶
- Debanjana Kar, Leopold Böss, Dacia Braca, Sebastian Dennerlein, Nina Hubig, Philipp Wintersberger, and Yufang Hou. (2025). MathBuddy: A Multimodal System for Affective Math Tutoring. https://arxiv.org/abs/2508.19993
Page metadata
| Field | Value |
|---|---|
type |
product |
id |
annotated-student-utterance-emotion-dataset-224-turns-for-text-based-emotion-recognition-i |
title |
Annotated student-utterance emotion dataset (224 turns) for text-based emotion recognition in math tutoring |
description |
Debanjana Kar et al.'s dataset contains 224 manually annotated MathDial-Bridge student turns with boredom, engagement, or neutral labels and polarity scores for developing emotion-recognition models in math tutoring dialogues. |
status |
review |
product_kind |
dataset |
generated |
by process:settle-candidates on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.