[4Rin1] Interactive 2

[4Rin1-41] Developing a Classroom Robot Service by Construncting Hierarchical Multi-Modal Rule Base

A Case Study on English Classroom

Kouya Takahashi¹, Momoko Akimoto¹, 〇Hidenao Abe¹ (1.Bunkyo University)

Keywords:Multi-modal Rule Base, Education Support Robot Service, Classroom AI

In recent years, cloud services that handle various forms of information, such as image recognition and speech recognition, have been developed using techniques based on machine learning algorithms such as deep learning. On the other hand, in order to more properly reproduce the behavior based on the professional knowledge of teachers in classroom, the development of a robot service that supports business tasks by flexibly adapting the motion to the user's situation is underway. However, in order to implement business knowledge collected from experts as robot services, it is necessary to build a rule base that provides flexible responses, subject to the results of processing various forms of data such as images, voice, and text. Therefore, in this paper, we describe a rule base construction for dialogue practice in English at junior high schools as a case study, by taking a hierarchical approach from expert opinion to the multi-modal rule-base implementation.