#PLENK2010 and Twitter

So I belong to 1000-somthing participants of #PLENK2010, which is a MOOC (massive open online course) run by Stephen Downes, George Siemens, Dave Cormier & Rita Kop. For more information view the course home page .

One of the participants, Tony Hirst, used the Gephi modularity statistic to cluster his Twitter network depending on the strengths of connections between members of that network. The Gephi modularity tool identified a number of different clusters. I found my Twitter ID, mediendidaktik, in the third cluster. It’s really interesting to see that the people I am connected with in #PLENK2010 are mostly (or all?) German speaking.  Also I seem to be having 9 connections, some of which I was not aware of, like Uni_Moodle or twoflower29.

PLENK2010 twitter cluster

Also compared to other clusters, “our” German cluster is not as densly populated as cluster 1 (Canada) or 2 (UK).  So again, are our connections in Twitter primarily based on the country we are based in?

3 thoughts on “#PLENK2010 and Twitter

  1. THe network I constructed was actually based on a Twitter list that contains people who have used the PLENK2010 hashtag more than a certain number of times on Twitter [ http://twitter.com/ousefulapi/plenk2010 ].

    For each person in the list, I grabbed their friends list (that is, the list of people they follow) and then identify which of their friends are on the PLENK2010 twitter list. I record all these friendships and that forms the basis of the PLENK2010 twitter network that I then visualise in Gephi.

    The Gephi modularity measure then looks for clusters of people who seem to be tightly connected together and “colours” each person accordingly.

    I then filtered the network on each of the clusters Gephi identified in order to generate the word cloud and location cloud.

    As the course goes on and folk make friends with each other, the network is likely to become more highly interconnected and the partitions we see at the moment may start to break down.

  2. Maybe over time the clusters will start to form around shared interests within the PLENK course. At the moment the main thing that binds us together in Tony’s analysis is the #PLENK2010 hashtag. Our follower networks are probably pretty much as they were when we started. But in the spirit of Anglo-German cooperation, I’m following you now😉

    What I’m intrigued about is how Tony gets at the lists of twitter people (I think he uses TwapperKeeper as the source) but I looked earlier and couldn’t find a post that explained how he extracts the data.

    • Hi Chris, indeed this should be interesting to see how clusters change over time based on what category – will it be still a country or will it become a topic? Are there any other categories that may be used?
      Happy about your spirti of Anglo-German cooperation! Ditto🙂 Ilona

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s