Chair: Naoyuki Kubota
Title: Human-Oriented AIoT Application
Department of Computer Science and Information Engineering
Tokyo Metropolitan University, Japan
Talk Abstract：
Various sensors are used around our daily life as multi-modal data, conventional recognition approaches deal with each sensor data respectively for recognizing human’s activity. Considering a human internal state, each modal closely related mutually, so this talk introduces how to recognize human behaviors according to individual differences of human personality. Firstly, a multi characteristic model architecture that combines the personalized machine learning models and utilizes each model’s prediction score in the inference is introduced. This architecture formed with reference to ensemble machine learning architecture. Secondly talks about robot expression based on individual traits adapting parameters thus shows different expressions to satisfy certain types of interlocutors. Our approach researches viewpoint from interaction in the mutual process, significant behavior explores the corresponding relationship between individual behaviors. Finally, I will talk about a comprehensive interaction robot application using a heart rate sensor and a brain-computer interface considering the internal state in the case of chatting towards co-learning.
Vita：
Eri Sato-Shimokawara (S'04–M'07) received the B.E., M.E., and D.E. degrees in systems engineering science from the Tokyo Metropolitan Institute of Technology, Tokyo, Japan, in 2002, 2004, and 2007, respectively.,She was a Research Fellow of Japan Society for the Promotion of Science from 2004 to 2007. She has been an Assistant Professor with the Faculty of System Design, Tokyo Metropolitan University, Hachioji, Japan, since 2007. Her current research interests include human–machine interactions, multimodal interactions, soft computing, and intelligent robotics.,Dr. Sato-Shimokawara is a member of the Institute of Electronics Information and Communication Engineers, the Japan Society for Fuzzy Theory and Intelligent Information, and the Japanese Society for Artificial Intelligence.
Plenary Talk：