DIGITAL LIBRARY
TALKYTUTOR: AN INNOVATIVE APPROACH FOR CHATBOTS FOR EDUCATION
Politecnico di Milano (ITALY)
About this paper:
Appears in: INTED2023 Proceedings
Publication year: 2023
Pages: 3088-3093
ISBN: 978-84-09-49026-4
ISSN: 2340-1079
doi: 10.21125/inted.2023.0857
Conference name: 17th International Technology, Education and Development Conference
Dates: 6-8 March, 2023
Location: Valencia, Spain
Abstract:
This paper proposes an original approach for supporting different educational actors using chatbot-mediated learning. An original technology (aCHAT) allows non-technical actors (e.g., teachers) to customize the learning experience, including content and conversational aspects.

Chatbots act as tutors, adaptively supporting each learner through (possibly complex) bodies of interactive content which is created independently from the chatbot. The underlying technology extensively uses the configuration to reduce time and effort in delivering a chatbot by empowering non-technical actors (i.e., publishers, authors, and teachers) to contribute directly.

Content is organized independently from the conversation: it can be created by authors and customized by teachers without the need for ICT support. It is organized into small learning items, using different topologies for different learning experiences.

The conversation is not implemented via coding but instead designed and configured via several devices: this allows easy customization and maintenance (basically configuring tables). Extensive configuration use rather than relying on built-in features provide many advantages. For example, less time to deliver, reduced costs and empowered various involved actors. Non-technical actors can directly adjust these relevant features to shape different learning experiences for the final learners.

A large experiment with teachers and students in higher education and junior high school has shown the viability and potential effectiveness of the approach. Students were highly pleased with the conversational interface and the possibility of controlling conversational features (e.g., loquacity and wording style). Teachers were pleased by the involvement of students and by the possibility of customizing the learning experience.

We did confirm our hypothesis that conversational interfaces can create a comfort zone for learning: “I’m not alone” (quote from a 13- years old student)
Keywords:
Chatbot, Adaptive Conversation, Learning Process, Technology-Mediated Learning.