The Challenges in Analyzing Online Videoconferencing Meetings
The research work crosses over cognitive science to human computer interaction. The work is focused on analyzing public online meetings ranging from peer to peer, to European Union meetings carried out on the Flashmeeting videoconference. The work is a part of the OpenLearn project at Open University. The aim of the research is to design a tool function that projects summary of what the conversation is about at every turn taking through video animation. The goal is to facilitate students to know which public Flashmeetings to select for their learning purposes. This is due to the fact that students have access to thousands of public Flashmeetings online. Moreover, each meeting may lasts up to 1 hour, making it more challenging for them to select. In order to achieve our goal is to begin constructing an automated analysis that can help categorize communications into different contexts (e.g., animation, sharing experience, problem solving). Starting from this point, we were then faced with a simple question: how do people readily recognize that this context of communication is, for example about animation? Thus, we needed to understand (i) how do people differentiate contexts of communications? And further on (ii) how do we embed those understanding in the conversation annotation so that a learning algorithm can induce patterns from those annotations? As a solution, we propose to combine conversation analysis, and discourse theory linked to the theory of learning and memory for analyzing the conversations. Thus, the proposal is focused on discussing the analyses through relating the observed conversations back to theories and the challenges we face to analyze those online meetings as well as automating some of the analyses.
Keywords: Conversation Analysis, Discourse Theory, Human Computer Interaction, Online Meetings, Memory, Learning
Dr. Nik Nailah Binti Abdullah
Researcher, Information Systems Architecture Research Division