I noticed this article in the New York Times.
Peter Sheridan Dodds and Christopher M. Danforth, a pair of statisticians from the University of Vermont, are hoping to harness the stream of messages flowing through the popular microblogging platform at any given moment to read public opinion and sentiment in real time.
The goal is to establish an index, akin to the Dow Jones industrial average, that can â€œgive an overall sense of how a collective body of people are feeling at any given point in time,â€ Mr. Dodds said.
Lately, there have been lots of initiatives that relate to this, often using Twitter as a source. Reading the description of this hedonimeter “one happy bird“, I take it this is one of the more serious initiatives that can be quite promising. What I like is the connection they make between repeated words such as ‘party’ or ‘office’ and relate it to an emotional state or mood:
The preliminary findings show what you would expect: Twitterers with more followers tend to be happier, and the number of positive tweets spikes around weekends (when words like â€œpartyâ€ are especially common). The mood drops during the week, coinciding with the rise of words such as â€œoffice.â€
At the same time, this is also one of my concerns with the tool. Which emotions are specifically measured and on what bases are the connections with stimuli being identified and determined?