Framing Early Alert of Struggling Students as an Anomaly Detection Problem: An Exploration

Eitel Lauría

Abstract

This exploratory study analyses the feasibility of implementing an early-alert system of academically vulnerable students using anomaly detection techniques for cases in which the number of struggling students is small in comparison to the total student population. The paper focuses on a semi-supervised approach to anomaly detection where a first stage made up of an ensemble of unsupervised anomaly detectors contributes features to a second-stage binary classifier. Experiments are carried out using several semesters of college data to compare the predictive performance of this semi-supervised approach relative to stand-alone classification-based methods.

Download


Paper Citation


in Harvard Style

Lauría E. (2021). Framing Early Alert of Struggling Students as an Anomaly Detection Problem: An Exploration. In Proceedings of the 13th International Conference on Computer Supported Education - Volume 1: CSEDU, ISBN 978-989-758-502-9, pages 26-35. DOI: 10.5220/0010471900260035


in Bibtex Style

@conference{csedu21,
author={Eitel Lauría},
title={Framing Early Alert of Struggling Students as an Anomaly Detection Problem: An Exploration},
booktitle={Proceedings of the 13th International Conference on Computer Supported Education - Volume 1: CSEDU,},
year={2021},
pages={26-35},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010471900260035},
isbn={978-989-758-502-9},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Computer Supported Education - Volume 1: CSEDU,
TI - Framing Early Alert of Struggling Students as an Anomaly Detection Problem: An Exploration
SN - 978-989-758-502-9
AU - Lauría E.
PY - 2021
SP - 26
EP - 35
DO - 10.5220/0010471900260035