Do you tweet the word "pleasee"—as in, please with an extra "e" on the end? Then odds are good you're a fan of The Bieb—aka Justin Bieber. A team of mathematical biologists at Royal Holloway, University of London, analyzed 75 million tweets from 250,000 users. Their analysis located several tight-knit communities that tweet far more heavily to each other than to the rest of Twitter and revealed a slew of communities in which members shared a common, special lingo  (see image) , they reported online last month in EPJ Data Science. One group focused on animal welfare was awash with puns, such as "anipals" or "pawsome." Technology-minded teachers employed terms like "edublogs." By using each tweeter's unique code words, the researchers were able to correctly predict their chosen communities—whether they be conservative Americans or college students fond of the Milwaukee coffee shop Alterra—close to 80% of the time, suggesting the words help tweeters identify themselves as community members.
See more ScienceShots .