In a new study published in EPJ Data Science, scientists from Royal Holloway University in collaboration with Princeton University delved inside the formation of tribe-like communities on Twitter.
The team found that language is an extremely reliable indicator of what communities users belong to, and vice versa – language used in tweets is an accurate indicator of what users might belong to that Twitter community.
Let’s take a closer look.
The EPJ Data Science study illustrates the tie between language on Twitter – in the sense of grammar, spelling and word choice, not tongue – and Twitter “tribes,” or communities. A “tribe” would be, say, Justin Bieber fans, or people in the social media industry, or authors.
Professor Vincent Jansen from Royal Holloway told Phys.org, “Interestingly, just as people have varying regional accents, we also found that communities would misspell words in different ways. The Justin Bieber fans have a habit of ending words in ‘ee’, as in ‘pleasee’, while school teachers tend to use long words.”
So, the research suggests that in an analysis of Twitterers who often tweet words that end in “ee,” about 80% of those people would be vocal Justin Bieber fans.
In the above figure, the top word given for each community is the most significant one in that community. Circles represent communities, with the area of the circle proportional to the number of users (>250 shown).
The widths of the lines between circles represent the numbers of messages (>5,000 shown) between or within community, and the colors of the self-loops represent the proportion of messages that are within users from that group.
The team even went so far as to produce a map of the communities they found on Twitter, showing how they have vocations, politics, ethnicities, and hobbies in common. A few “tribes” of note: an “anipals” group interested in hosting parties to raise funds for animal welfare, and a growing community interested in the concept of gratitude.
Here’s the main takeaway: as social media enables us to communicate with almost anyone across the globe, we’re using it to communicate mostly with people just like ourselves.