Can a Twitter 'happiness measure' work for agencies?
How happy is the city or state in which you live? Researchers at the Department of Mathematics and Statistics at the University of Vermont recently released a study of geotagged Twitter feeds broken down by geographic location, rating how happy or sad the tweets were. The researchers rated the mood of each geotagged tweet made in 2011 based on how certain words correlate with the geographic area and the tweeter’s presumed state of happiness.
Anyway, although this seems just a cute little study about Twitter, there are big possibilities for public-sector agencies in tying this methodology to data from other sources, such as the Census Bureau. The Office of Naval Research, for example, is using Aptima's Epidemiological Modeling of the Evolution of Messages (E-MEME) system to develop ways of tracking public sentiment on social media, in order to predict uprisings or protests.
Police use social media to help track gang and other criminal activity, and monitoring of social networks is being used for everything from predicting the stock market to tracking the spread of flu.
The methods used at the University of Vermont could be applied to track public response to agency services, government policies or economic conditions as another channel of feedback.
Of course, social media tracking has its limits. In this study, for instance, any tweets that were not geotagged could not be included. So, if you are worried about your 140-character treatise ending up in a study like this, just disable your geotagging. But if you are worried about things like that, maybe you are just in a really sad geographic area.
Posted by Greg Crowe on Feb 26, 2013 at 10:42 AM