University of Twente Student Theses
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.
AI-Driven Personalization in Educational Dialogues : Microtool for Child Profile Building
Sandu, M.A. (2025) AI-Driven Personalization in Educational Dialogues : Microtool for Child Profile Building.
|
PDF
2MB |
| Abstract: | Robots have yet to reach the level of personalization necessary to foster long-term interactions with children. Existing systems struggle to establish the needed social connections as they depend on scripted content and rigid dialogue structures, not adjusting to individual users. To address this gap, this research implements a chatbot tool that, using a knowledge graph (KG) and strict prompt rules, creates an individualized conversational experience. It references each child's past data, personalizing questions and transitions throughout a structured process while filling 29 fields in a user model. The chatbot was tested in a within-subjects study where children aged between 9 and 11 interacted with both the chatbot and a personalized paper-based letter with integrated questions. Results show that the chatbot significantly increased data completeness, with the missing fields reduced by over 90% compared to the baseline, leading to more updated information, and was rated as more engaging by children. Qualitative feedback highlighted a preference for the chatbot’s interactive and conversational style, though both modalities were seen as personal in different ways. These findings support the use of AI-driven, memory-based interfaces for building rapport and gathering personalized data in child-robot interaction contexts. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/107276 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page