Beyond equality: feature extraction and detection for classification of answers
Teaching Mathematics and Its Applications · 19.03.2026
DOI: 10.1093/teamat/hrag005 ↗Kurz gefasst
Fachartikel in der Zeitschrift Teaching Mathematics and Its Applications, online erschienen am 19.03.2026. Verfasst von Matti Harjula. Die Arbeit nennt 12 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Testing for equality is a key feature in computer aided assessment of mathematical sciences. However, it has its problems when dealing with partially correct answers and feedback generation. There are also cases where some types of equality might be overly sensitive or permissive. This paper explores non-equality based classification of answers, based on feature extraction, and presents a test case question requiring it. This test case can be used to test the capabilities of both automated assessment systems and authors using them.
Zitieren
APA
Harjula, M. (2026). Beyond equality: feature extraction and detection for classification of answers. Teaching Mathematics and Its Applications. https://doi.org/10.1093/teamat/hrag005
Vancouver
Harjula M. Beyond equality: feature extraction and detection for classification of answers. Teaching Mathematics and Its Applications. 2026. doi:10.1093/teamat/hrag005
BibTeX
@article{Harjula2026,
author = {Harjula, Matti},
title = {Beyond equality: feature extraction and detection for classification of answers},
journal = {Teaching Mathematics and Its Applications},
year = {2026},
doi = {10.1093/teamat/hrag005}
}
EUVIMED