Designing a Mobile Digital Backchannel System for Monitoring Sentiments and Emotions in Large Lectures

Designing a Mobile Digital Backchannel System for

Monitoring Sentiments and Emotions in Large Lectures

By Peerumporn Jiranantanagorn

PhD Candidate School of Computer Science, Mathematics and Engineering

Flinders University

Outline • Introduction • Research questions • Related work • Overview of the system • User interfaces • Technical implementation • Conclusion and future work

Introduction • In a large lecture, it is difficult for lecturers to process,

respond and know overall emotions and sentiments of students in real time while they are teaching.

Introduction • One way to make a large classroom more

manageable and engaging is to use a digital backchannel system.

• However, the scattered and sparse nature of posts makes it impossible for the lecturer to get a current overall picture of students’ learning as well as the emotions and sentiments of students in a large lecture.

Research questions • How to design and develop a system to support

lecturer to know students’ real-time morale and the current important discussions during her/his lecture?

• How to design and develop a system with a microblogging user interface that allows students to express their sentiments and emotions in a large lecture?

• How to design a questionnaire to evaluate the user acceptance and user satisfaction of the system from the perspectives of both the lecturers and the students?

Related Work • A comparison of the existing digital backchannel

systems and our system Backchannel

Systems Microblogging

Support

Post Classification Vote Sentiment Emotion

Hotseat (2010) n/a n/a Backstage (2011) n/a n/a ClasCommons (2012) n/a n/a n/a ActiveClass (2003) n/a n/a n/a ClasSense

Overview of the ClasSense

Mobile Application for Student

Web Application for Lecturer

Technical Implementation • The ClasSense mobile and web applications have

been developing using jQuery framework, JavaScript, PHP and MySQL.

• All applications are hosted in the cloud. • Emotion expression is currently through emoticons

and selecting from Kort’s twelve learning relevant emotions hashtags (Kort 2001), which are “#frustration”, “#disappointment”, “#confusion”, “#satisfaction”, “#hopefulness”, “#confident”, “#dispirited”, “#boredom”, “#dissatisfied”, “#interest”, “#curiosity”, and “#enthusiastic”

Technical Implementation • Morale score for plotting graph is based on

normalising values from SentiStrength score, number of posts and range of score (1…5).

• For web application, post ranking is based on morale scores, number of likes and dislikes, number of comments and post time.

• For mobile application, post ranking is based on ageing score, number of vote and number of comment.

Evaluation

• The system will be evaluated using questions framed with the – Technology Acceptance Model and – Seven Principles for Good Practice in

Undergraduate Education • Still researching on a Usability testing of

the system

Conclusion and Future Work • The ClasSense system has been developed to help

lecturer monitor the morale of students and respond to the important issues students have in real-time.

• Future work includes – System stability and validity testing – Customise and test the SentiStrength – Pilot and formal evaluation in large lectures

Thanks for listening