Towards Automatic Recognition of Perceived Level of Understanding on Online Lectures using Earables
Dongwoo Kim, Chulhong Min, Seungwoo Kang · 2021
The COVID-19 pandemic has seriously impacted education and forced the whole education system to shift to online learning. Such a transition has been readily made by virtue of today’s Internet technology and infrastructure, but online learning also has limitations compared to traditional face-to-face lectures. One of the biggest hurdles is that it is challenging for teachers to instantly keep track of students’ learning status. In this paper, we envision earables as an opportunity to automatically estimate learner’s understanding of learning material for effective learning and teaching, e.g., to pinpoint the part for which learners need to put more effort to understand. To this end, we conduct a small-scale exploratory study with 8 participants for 24 lectures in total and investigate learner’s behavioral characteristics that indicate the level of understanding. We demonstrate that those behaviors can be captured from a motion signal on earables. We discuss challenges that need to be further addressed to realize our vision.