I spent Thanskgiving and the day after relaxing in my own peculiar way--by mining the Wikileaks 9/11 pager data.
Here are some early results:
I started by pulling out all the email addresses from the pages and storing them in their own table, with keys to their original page. I also pulled the unique pager numbers from the pages. What I got was a bipartite directed graph with one side being emails and the other being pagers, with messages functioning as edges. Using Django, Graphviz, Cinelerra and a bunch of other tools, I was able to make a video of the graph as it lights up on each relevant page.
Finally coming up for air and getting a little time to blog about some interesting stuff I've been a part of. Just to get started, I wanted to show off a video my colleague Claudio Midolo shot of our fraction game in SMALLab, which we've provisionally dubbed "FracAttack".
I've put together a .csv (comma-separated file) with the results I pulled off of Twitter for the first 48 hours of the Iranian election events. Be aware--it's about 20MB. Hopefully, many of you will find this useful in your own research. The columns are tweet id, date and time, text, profile image path, twitter username, twitter user id, and twitter user id of the immediate "reply to" (note that, in my graph analysis, I keep track of all @'s in the message, not just the first one as Twitter does. Only that first id is listed in the data file.)
In the NYC Learning Network projects, we've been doing a lot recently with using a webapp on an iPhone (here's our first stab at a search engine using Eric Sanderson's brilliant Mannahatta data). There's all kinds of ways to do this (we're most excited by PhoneGap). For me, as a Ubuntu Linux user, this is pretty great--I can help develop large pieces of the application without dealing with Apple and its locked-down way of doing things. There's just one problem--I really need to see how it works with WebKit, the engine behind Safari and other browsers.
If you've read my swineflu analysis, some of this should make sense. I ran a search on '#iranelection OR Tehran OR Ahmadinejad OR Mousavi' in Twitter for the period between Friday and Sunday evening. From the 79,957 results I got back, below is some graph analysis of what came out.
Here is the latest in my continuing series on analyzing Twitter conference backchannels by their hashtags and replies/retweets. This one, though, is a bit different and special... because I was actually at the conference! Below is my breakdown of Games + Learning + Society 2009 via the #gls and #gls09 hashtags.
Pretty excited to be in Madison--it's a really beautiful town. I'm at GLS 09 for the week to check out the latest and greatest in my little field. I'll be presenting with Dave Birchfield and Katie tonight for the SMALLab poster session, and I'm running a game with Colleen, Eric Zimmerman, and John Sharp using twitter called BACKCHATTER which should be a lot of fun to play.
If you're around, I'm the tall scruffy one in the lime green shirt!
Because I've recently been... let's just come out and say obsessed with looking at the social relationships that seem to emerge from examining sociograms of Twitter users within the "channel" of a particular hashtag, here's another one I thought was interesting: Media in Transition 6, a.k.a. #mit6.
Just a quick post about another conference's Twitter backchannel I analyzed recently. Take a look at my posts on #swineflu and #09ntc to get a full picture of what I'm up to here. Basically, I'm looking at the network formed by replies and retweets in Twitter inside of a particular hashtag. Here, I'll go over the results of Museums and the Web 2009, a.k.a. #mw2009.
I just did a run on the first two days of the 2009 Nonprofit Technology Conference using the tools I've been working on (see my post on #swineflu earlier this week.) Using the hashtag #09ntc, I parsed 3834 tweets, and I looked up the hubs and authorities, plus generated the graph of the largest strongly connected component within the larger directed graph created from all the "@" replies and retweets.
Copyright Mike Edwards 2006-2009. All content available under the Creative Commons Attribution ShareAlike license, unless otherwise noted.