Comments on: CCK08 – Chat Interaction frequency

By: Viplav Baxi

Viplav Baxi — Thu, 18 Sep 2008 03:47:23 +0000

Hi George,
I saw a bit of taxonomy around the analysis in Maarten’s paper. The challenge is to design a system that can plug in to popular conferencing system and collaboration tools that are based on a semantic analysis and classification of contributions with minimal human intervention. Such a system could evolve within a 2.0 collaborative folksonomical context.
Applications are huge. I would love to get with people who would be interested in developing something like this. A weak area, in fact an area where I have no expertise, is that of semantic analysis for me though .

By: gsiemens

gsiemens — Wed, 17 Sep 2008 12:54:45 +0000

Hi Viplav – the goal here, of course, is to get our software to have analysis tools built right in. Education metrics – or a similar term – are important in helping us to understand how we are contributing in relation to others, who the dominant participants are, etc. What you have created is a diagram of participation at a basic level. Once we are better able to not only detail who contributed but also the manner of their contribution (content analysis), we can then begin to get a better sense of overall value.
George

By: pingadohtor

pingadohtor — Thu, 11 Sep 2008 22:31:47 +0000

heh, glad to hear we are on the same wavelength.

will be interested to see what else you do with the statistical analysis of the course.

thanks!

By: Viplav Baxi

Viplav Baxi — Thu, 11 Sep 2008 18:33:19 +0000

Yes. I started down that path but did not have a formal taxonomy yet that could cover the interaction types. And then getting into the threads – yes, that was a thought as well. I will do it the moment I get some more time.

By: pingadohtor

pingadohtor — Thu, 11 Sep 2008 13:51:48 +0000

would be interested to see if you can analyse the greetings, ie “hi” “how are you” volume, as well as who are the community builders (responders to hi/how are you)

also, would be interesting to see the stats for content related posts