Temporal Sentiment Analysis of Learners: Public Versus Private Social Media Communication Channels in a Women-in-Tech Conversion Course

Jialin Yu,Olanrewaju Tahir Aduragba,Zhongtian Sun,Sue Black,Craig D. Stewart,Lei Shi,Alexandra I. Cristea

DOI: https://doi.org/10.1109/iccse49874.2020.9201631

2020-01-01

Abstract:Social media is ubiquitous, a continuous part of our daily lives; it offers new ways of communication. This is especially crucial in education, where various online systems make use of (perceived) public or private communication, as a means to support the learning process, often in real-time. However, not much research has been carried out in analysing and comparing such channels and the way participants use them. Thus, this paper analyses a course offering both public and private types of communication to its participants. Participants communicate via two social media channels (beyond traditional email etc.): Twitter (open to the public) and Microsoft Teams (for internal communication). In this paper, we specifically analyse the communication patterns of learners, focusing on the temporal analysis of their sentiments on the public versus the private platform. The comparison shows that, as possibly expected, there exist similarities between expressed sentiment in public and private channels. Interestingly however, the private platform is more likely to be used for negative utterances. It also shows that sentiment can be clearly connected to events in the course (e.g., the residentials increase both volume and positivity of comments). Finally, we propose new measures for sentiment analysis to better express the nature of change and speed of change of the sentiment in the two channels used by our learners during their learning process.