The Facebook Data Team has published an analysis of the status updates of Facebook users, by categorizing words according to the 68 categories of the Linguistic Inquiry and Word Count Dictionary, and tabulating the frequencies of their use. It's fairly interesting to see this kind of analysis applied to Facebook, but unfortunately doesn't reveal much in the way of "A-ha!" insights. For example, this chart of the frequency of six categories of words expressed throughout the day shows pretty much what you'd expect:
(I do wonder whether "Hour of Day" is in local time or all mapped to the same timezone, though: I'd expect the variation to be more pronounced if the analysis were based on local time, and if not I'd have to assume this represents US-based Facebook users only.)
Notable for R users is that all of the charts in the Facebook post were clearly created with Hadley Wickham's ggplot2 package. If you haven't tried ggplot2 yet, check out the ggplot2 website for resources for creating beautiful graphics with R. If you're already a ggplot2 user, you might have missed that it was updated just before Christmas with improvements to legends and axes and several bug fixes. Read the update announcement for the full details.
Facebook Data Team: What’s on your mind?
Cool idea! Source code?
Posted by: Asjudc | December 30, 2010 at 02:18
You'll have to ask the Facebook Data Team for the source code, but I doubt they'd be willing to share the source data.
Posted by: David Smith | December 30, 2010 at 11:26