Sentiment analysis and visualisation in a backchannel system

Abstract

Digital backchannel systems have been proven useful to help a lecturer gather real-time online feedback from students in a lecture environment. However, the large number of posts made during a lecture creates a major hurdle for the lecturer to promptly analyse them and take actions accordingly in time. To tackle this problem, we propose a solution that analyses the sentiment of students' feedback and visualises the morale trend of the student population to the lecturer in real time. In this paper, we present the user interface for morale visualisation and playback of ranked posts as well as the techniques for sentiment analysis and morale computation.