March 14, 2013
Research Shows Twitter Users Band Together In ‘Tribes’
Alan McStravick for redOrbit.com — Your Universe Online
When one hears the word ℠tribe´, an image of a more primitive collection of people gathered into a loose society is often evoked. Here in the US, the mind leaps to our own indigenous population of Native Americans. However, especially when taking into account the actual definition of the word, our misconceptions come off as too simplistic. Tribes and their activities, it turns out, are not relics of a bygone era. And a new collaborative project between researchers at the Royal Holloway and Princeton Universities aims to detail the relevance of a tribe as it pertains to social media.
The definition of a tribe is far broader than many might even consider. They are comprised of groups of people who are connected to an idea, leader and each other. The evidence based study by researchers sought to explore how people have formed into tribe-like communities on social networking sites such as Twitter.
The idea of a tribal mentality online is not new news for Out:Think owner Tim Grahl. In the Think First blog for his site, meant to speak to authors hoping to engage readers and increase personal following, he posits the question, “When people join your tribe, what are they becoming a part of? What is the bigger “idea” behind your work?” Thanks to this most recent study, Mr. Grahl now has documented proof to bolster his assertion of social media tribal formation.
The paper, entitled "Word usage mirrors community structure in the online social network Twitter" was published in EPJ Data Science, and in it, they detail how they found these online communities have a common character, occupation or interest. In many, the development of a unique and distinctive language was even recognized.
"This means that by looking at the language someone uses, it is possible to predict which community he or she is likely to belong to, with up to 80% accuracy," according to Dr. John Bryden from the School of Biological Sciences at Royal Holloway. "We searched for unusual words that are used a lot by one community, but relatively infrequently by the others. For example, one community often mentioned Justin Bieber, while another talked about President Obama."
A colleague at the university and on the study, professor Vincent Jansen, added, “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.”
In the process of their research, the team devised a map of the online communities. This map detailed how individual communities had vocations, political views, ethnicities and hobbies in common. The map was able to be created due to the team focusing on the sending of publicly available messages through Twitter. This method allowed the team to record conversations between two or many participants.
Once the initial data was collected, the team grouped the users into the mapped communities through the use of cutting-edge algorithms from both physics and network sciences. Specifically, the algorithms sought to identify individuals that tend to send their messages to other members of the same community.
With the team being able to identify and map the social communities, Dr. Bryden made the suggestion that they analyze the language use within each of the communities.
Dr. Sebastian Funk, a study co-author from Princeton University, stated, “When we started to apply John´s ideas, surprising groups started to emerge that we weren´t expecting. One ℠anipals´ group was interested in hosting parties to raise funds for animal welfare, while another was a fascinating growing community interested in the concept of gratitude.”
This innovative and forward thinking research utilized many data collection techniques not used previously. For this reason, the Royal Holloway University has filed a patent application to safeguard their future use.