| 69302 |
Creator |
3a7e1208601d2480ab76f5a4888d255f |
| 69302 |
Creator |
c79cc1129e553cdca30fa856443ac46e |
| 69302 |
Creator |
974911e8c683e404920182bccae28818 |
| 69302 |
Date |
2020 |
| 69302 |
Is Part Of |
repository |
| 69302 |
abstract |
The poster focuses on a recommender method that is tightly related to predictive learning
analytics in distance higher education focused on the identification of students at
risk of not submitting their assignments and subsequently failing their courses. Given
a lack of student time to the assignment deadline, the method aims to provide a minimalistic
recommendation for students to increase their chances of submitting the assignment
so that they survive a possible difficulty they encounter. We formally define the
task as an optimisation problem and propose a simple algorithm that will serve as
a baseline for further improvement. On an offline evaluation on one STEM course, taking
only students predicted as at-risk, those that followed the recommendations were associated
with higher submission rates than if they only accessed any online resource. |
| 69302 |
authorList |
authors |
| 69302 |
presentedAt |
ext-742e1268b1a8935efad30f70bc2ffdeb |
| 69302 |
status |
peerReviewed |
| 69302 |
type |
AcademicArticle |
| 69302 |
type |
Article |
| 69302 |
label |
Hlosta, Martin ; Bayer, Vaclav and Zdrahal, Zdenek (2020). Mini Survival Kit: Prediction
based recommender to help students escape their critical situation in online courses.
In: Proceedings of the 10th International Conference on Learning Analytics and Knowledge
(LAK20), 23-27 Mar 2020, Frankfurt am Main, Germany. |
| 69302 |
Title |
Mini Survival Kit: Prediction based recommender to help students escape their critical
situation in online courses |
| 69302 |
in dataset |
oro |