***** To join INSNA, visit http://www.insna.org *****
Take a look at this network I built from a Twitter discussion -- some interactions are with the group, as a whole, and some are with specific others...
You can hide the "group node" and just see the remaining links between discussion members, take network measures at either point.
On Mar 31, 2011, at 4:48 AM, Gabor Majdan wrote:
> ***** To join INSNA, visit http://www.insna.org *****
> Dear all
> I have a big dataset of posts from an online community (discussion/help forum) and I want to analyze the interactions between users based on the posts they have written.
> Every post is in a thread, however, the posts are not hierarchical; a message posted to a thread doesn't show whom the author replied to, and all messages are on the same level (i.e. no indentation). Considering the simplest ways, this would leave me three choices as for network building: create a complete graph out of a thread (connect everyone with everyone), create a star-graph with the thread-starter in the middle or connect users to the previous person who replied.
> Do you happen to know papers on building such an interaction network out of non-hierarchical discussions, but not with the above mentioned methods?
> Thank you
> Gabor Majdan
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.