Conceptualization of IMS that Estimates Learners' Mental States from Learners' Physiological Information Using Deep Neural Network Algorithm

Tatsunori Matsui, Yoshimasa Tawatsuji, Siyuan Fang, Tatsuro Uno

Abstract: To improve the efficiency of teaching and learning, it is substantially important to know learners’ mental states during their learning processes. In this study, we tried to extract the relationships between the learner’s mental states and the learner’s physiological information complemented by the teacher’s speech acts using machine learning. The results of the system simulation showed that the system could estimate the learner’s mental states in high accuracy. Based on the construction of the system, we further discussed the concept of IMS and the necessary future work for IMS development.