Perceptual Recognition of States in Remote Classrooms
Ian Beaver, Akira Inoue · 2005
We present a method to recognize states in remote classrooms to provide autopilot services for distance education: no session, in-session, and question (i.e. a student in the classroom draws the instructor's attention). We study such a method that uses fuzzy classifiers to recognize above states and a simple feature space presenting a signal of human body movement. This computational model is largely inspired and justified by previous studies on a computational model of perception according to Gestalt theory. Mass assignment theory (MAT) is used for constructing and representing the mapping between a space of meaning (i.e. perception) and a space of signal (i.e. sensor). To show effectiveness, we conducted a comparative study between a conventional approach using fuzzy c-means algorithm and the method based on MAT.