Precedability Prediction Between Open Educational Resources - Archive ouverte HAL
Conference Papers Year : 2024

Precedability Prediction Between Open Educational Resources

Abstract

The abundance of Educational Resources (ERs) has allowed people to have access to a vast amount of knowledge. However, it can be difficult, for both educators and learners, to navigate through these resources. One way to facilitate navigation is to identify useful relations between these resources. This can improve the teaching and learning experiences by allowing the users to go from one resource to another based on the identified relations, such as precedence. In this work, we introduce the notion of precedability between educational resources; whether a resource A can precede another resource B. Then, we propose a two-step method to identify precedability relations between educational resources. Our method structures the educational resources in an enriched Knowledge Graph (KG). Then, it uses a Graph Neural Network (GNN) model to predict precedability relations. Our method performed better than multiple baselines on different benchmarks.
Fichier principal
Vignette du fichier
goodit24-55.pdf (659.47 Ko) Télécharger le fichier
Origin Publication funded by an institution
Licence

Dates and versions

hal-04654407 , version 1 (22-07-2024)

Licence

Identifiers

Cite

Aymen Bazouzi, Hoël Le Capitaine, Zoltan Miklos, Mickaël Foursov. Precedability Prediction Between Open Educational Resources. International Conference on Information Technology for Social Good (GoodIT ’24), Sep 2024, Bremen, Germany. ⟨10.1145/3677525.3678686⟩. ⟨hal-04654407⟩
169 View
94 Download

Altmetric

Share

More