The rapid growth of online, video-based teaching programmes has greatly reduced the amount of regular interaction between learners and educators, leading to a drop in engagement, a key element of educational success. At the same time, learners and educators are unfamiliar with how to replicate natural interaction while teaching and learning online. This problem is likely to affect all organisations that use virtual or blended programmes, from FE colleges to businesses.
To help address this increasingly pervasive problem, Conan Lab’s existing prototype uses AI/ML analytics to “invisibly assess” the quality of learner engagement during live video conferencing classes and feed this data back to the educator. This can help educators understand the quality of interaction during a class, identify strengths and weaknesses, and support their own learning and development.
This project will support the development of a minimum viable product to be deployed at scale over an extended period, providing large volumes of data and enabling identification of trends in learner engagement and feedback from test users.
Ultimately, it is hoped that the approach will result in a product able to predict the likelihood of engagement or disengagement for a given teaching approach and help improve educators’ interaction skills.