Period: 2018 ~ Present
Partner: Microsoft OneNote
In higher education, the popularity of flipped and blended classrooms has increased the number of students learning through asynchronous, online educational technologies. It is important that these technologies support students’ cognitive and social engagement. In my collaboration with UBC and Microsoft, I have developed an analytic framework that classifies students’ level of cognitive engagement through their multidimensional clickstream data using unsupervised machine learning methods (e.g., self-organized mapping).
- Analyzed clickstream big data using unsupervised machine learning methods (e.g., self-organized mapping)
- Designed an AI agent to identify the conceptually important aspects of a video and automatically generates relevant exam questions in order to test students’ understanding
- Tools: Deep learning
- Seo, K., Dodson, S., Harandi, N., Roberson, N., Sunani, S., Fels, S., & Roll, I. (submitting). How students learn with video: An analytic framework to unveil students’ cognitive engagement with video in different educational contexts. Computers & Education.
- Fong, M., Dodson, S., Harandi, N. M., Seo, K., Yoon, D., Roll, I., & Fels, S. (2019, June). Instructors Desire Student Activity, Literacy, and Video Quality Analytics to Improve Video-based Blended Courses. In Proceedings of the Sixth (2019) ACM Conference on Learning@ Scale (p. 7). ACM.
- Wu, F., Zhou, Q., Seo, K., Kashiwaqi, T., & Fels, S. (2019, March). I Got Your Point: An Investigation of Pointing Cues in a Spherical Fish Tank Virtual Reality Display. In 2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR) (pp. 1237-1238). IEEE.