A Method of Estimating Cooperative Activities in Collaborative Learning Based on Participants’ Spatial Relationships


Collaborative learning has become more and more important in education area. In most collaborative works, students are separated into groups, where the possible scope of teachers might be strongly limited. Therefore, automatic feedback based on sensing the state of students during collaborative work is helpful for effective educational guidance. In this work, we focus on the spatial relationships of each participant pair and tried to examine the feasibility of estimating learners__ cooperative activities. To achieve this purpose, KINECT is used to record space coordinates of learners during collaborative works. As the objective evaluation, average distances between students were calculated in every 10 seconds based on the recorded data. On the other hand, the subjective evaluation was performed by 6 researchers with monitoring the collaborative work video and giving a 5-grade mark for each pair of students in every 10 seconds. In order to verify the effectiveness of the spatial measurement, we calculated the correlation coefficient between objective and subjective evaluation within a one-minute time window, which is shifted by 10 seconds through the collaborative work (ca. 5 minutes). The result shows that the 1-minute time spans with correlation coefficients of 0.5-0.85 occupied around 47% (in average) of the whole collaborative work, during which students are cooperatively learning. This suggests that the spatial relationship is able to estimate the existence of cooperative activities between students, and is able to be used for online student state detection. In our future work, students' posture, gesture and verbal data should be also involved.

Author Information
Nuo Zhang, KDDI R&D Laboratories, Inc., Japan
Masami Suzuki, KDDI R&D Laboratories, Inc., Japan
Hiroaki Kimura, KDDI R&D Laboratories, Inc., Japan

Paper Information
Conference: IICTCHawaii2016
Stream: e-learning and collaborative learning

This paper is part of the IICTCHawaii2016 Conference Proceedings (View)
Full Paper
View / Download the full paper in a new tab/window

Comments & Feedback

Place a comment using your LinkedIn profile


Share on activity feed

Powered by WP LinkPress

Share this Research

Posted by James Alexander Gordon