Skip to main content

View Diary: High Impact Diary Lists: Current Methods (25 comments)

Comment Preferences

  •  trial and error (0+ / 0-)

    will certainly be involved.  But we have to start somewhere.  

    I don't have any other data about the users.  I'm not sure how we could go about getting it, or if we would want to.  We can categorize by activity; writes diaries, recommends diaries, comments in diaries, plus combinations.

    Why don't I just go ahead and send you a data set?

    I have in mind a single day, using uid since you don't seem to miss the names, using a pajek compatible format (I will test here), and coloring writers differently from readers (commenters or recommenders).  Yesterday there were 9441 links and 3099 nodes.  Average degree of nodes is ~3 and network density is 1%.

    It will take a bit to pull that together, if you agree that's the one to start with.

    If we used the last published week, there would be 54541 links and 6770 nodes. Connection  counts would range from 1 to 7.  Average degree of nodes is ~8 and network density is 0.7%.

    These graphs will likely be dominated by the high commenters and high recommenders.  Last week va dare gave out 576 recommendations, and there were 55 others who gave out 50 or more recommendations.  

    •  Sure send it along (0+ / 0-)

      when you have it ready and I will see what I can do.

      Sounds like a good place to start

      Republicans worry about our souls and their bellies. Democrats worry about their souls and our bellies

      by plf515 on Wed Jul 12, 2006 at 09:03:14 AM PDT

      [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site