Design and evaluation of an ontology for a conversational agent to support game-based independent learning in children.
Keywords
Loading...
Authors
Issue Date
2022-06-18
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
Teachers play an important role in the development of independent learning in children. Because of a shortage
of teachers in primary schools, and because of an increase of class size, teachers are nowadays not always able
to support all children with problems related to independent learning. Different studies have investigated
how the combination of educational games with the assistance of social robots can support children during
learning. Most studies, however, focus on supporting learning for a specific topic with a specific group
of children, such as children with additional needs. In this project, we investigated how we could design
and evaluate an ontology, which could be used by a conversational agent, that captures information on
general difficulties children face related to independent learning (defined as learning obstacles). We applied
a user-centered approach, in which we obtained knowledge on independent learning in children from online
interviews with teachers. This knowledge was interpreted and used to design an ontology. In an online
survey we evaluated the performance of the ontology based on learning obstacle identification and learning
obstacle mitigation with respect to knowledge of teachers. We found that the ontology is relatively capable
of identifying learning obstacles, but is especially good at indicating applicable strategies to apply. The
ontology is overall able to outperform randomly generated output. In an evaluation on structure, we found
that the ontology could be improved to be more complete and consistent. The ontology is, however, very
expandable and adaptable. We therefore conclude that this ontology provides a good starting point for the
creation of a more extensive and more complete formalization of the domain of independent learning. Such an
ontology will add to our understanding of independent learning in children and will enhance communication
between robots and humans.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen