April 27th, 2006
Most of us are familiar with the “consumer confidence” reports that tell us, at approximately monthly intervals, whether we’re collectively feeling more or less optimistic about the state of the world. One of the most widely reported metrics of such confidence is the Conference Board Consumer Confidence Index; it recently told us that, while we (Americans) were feeling reasonably cheery in March (a rating of 107.5, using the confidence level in 1985 as a “base” of 100), we’re now up to 109.6 in April. Indeed, that’s the most optimistic we’ve been since May of 2002, when the index reached 110.3.
Well, that’s all very interesting — but the next release of consumer confidence data isn’t scheduled to occur until May 30th. What if I want to know the level of public confidence tomorrow, or next week? And what if I’m interested in a mood other than just “confidence”? What if I’d like to know how many people are feeling aggravated, amused, angry, annoyed, anxious, bitchy, blah, busy, cranky, confused, crappy, creative, cynical, depressed, devious, distressed, ditzy, dorky, drunk, ecstatic, frustrated, geeky, horny, intimidated, irate, melancholy, naughty, scared, shocked, or worried? What if I’d like to know whether these moods have shifted appreciably in the past week, the past month, or the past year?
It turns out that such moods — as reflected by the moods of postings on the blogosphere — are available. I first noticed this a few days ago, in an article entitled “Software tracks mood swings of blogosphere,” which reported on a research program called MoodViews, which tracks the mood of roughly 10 million blogs hosted by LiveJournal. According to the article, there are roughly 250,000 new LiveJournal postings per day (an interesting statistic in itself: only 2.% of the blogs are updated each day), of which 150,000 have assigned one of several hundred labels for moods — e.g., moods like the ditzy-dorky-drunk labels listed above. MoodViews tracks these moods, and can show how they vary over time.
Sounds interesting, but I was too busy at the time to pursue the matter further. (Hmm, how many of us are busy at any given time? Are we busier now than we were earlier this year? Stay tuned for answers…) But this morning, I typed “MoodViews” into my Google search engine, and within seconds, I had navigated over to the MoodViews web site. Fascinating stuff! I don’t care how busy you are; push aside the distractions of the moment, and surf over there yourself, and learn about Moodgrapher (which tracks global mood levels), Moodteller (which predicts global mood levels), and Moodsignals (which helps in understanding the underlying reasons for mood changes). All of this is part of an ongoing research project at Amsterdam University in the Netherlands, and I suspect that the underlying tools will become more sophisticated as time goes on. The two primary researchers are Gilad Mishne and Natalie Glance, and they’ve published several papers on their work; you can see a list of the publications here.
The Moodgrapher tool, by the way, is accessible on the Moodviews website: you can choose any one of the hundreds of LiveJournal mood-tags, select a time period, and create your own graph right in front of your dazzled eyes. For example, are people feeling busier today than last week, last month, or earlier in the year? I selected the mood of “busy,” and the time period of January 1, 2006 to April 27th (this morning), and here’s what I got:

