Link Results
So for an initial summary of my results on tracking link flows:
I collected data from 10/25/2005 to 12/08/2006. Over that time I tracked 29,223 distinct links from 99,724 users, and detected 722,942 occurrences of one user posting a link and then a user who reads their blog posting the link.
Viewing this data as a connected graph there are 13,626 distinct pieces, however the largest contains 104,255 users and the second largest 56.
The links that constitute the majority of this data are those from internet memes such as Rate My Life which is the most transmitted link and was transmitted 22,747 times. This is excluding the three largest portal sites (which were completely filtered from this data) but several other portal sites are still present in the data and present a form of spam.
Oftentimes links will come bundled as can be seen from this user's post. The results of the online quiz this user took come along with several other links. The link to the portal site is somewhat undesireable to track as the portal will contain many different quizzes, confusing the data.
No effort was made to construct a model beyond a PageRank style analyzation of the graph.
