Communications of the ACM
A System Detects Global Trends in Social Networks Two Months in Advance
Using a set of randomly selected users (C), some followers are framed in the sensor-friends group (S). These occupy a more central position in the web and, therefore, find out about information and spread it before others.
Credit: Carlos III University of Madrid (Spain)
Researchers at Carlos III University of Madrid, the Autonomous University of Madrid, NICTA, the University of California, San Diego, and Yale University have developed a monitoring method that identifies what information will be relevant on social networks up to two months in advance.
The goal of the research was to test what is known as the sensors hypothesis on social networks, investigating if it is possible to find a group of people with a special position that would allow the information that goes viral globally on the Internet to be monitored.
The researchers analyzed a sample of data from 40 million Twitter users and 15 billion followers in 2009 and found that each user had an average of 25 followers, who in turn had an average of 422 followers. "This means that a person's followers have a role in a social network that makes them very relevant when it comes to spreading or receiving information," says Autonomous University's Manuel Garcia Herranz. He says the research showed that "sensor-friends" play a more important role than what was previously believed because they receive information long before the previously chosen users.
The researchers note that data from just 50,000 Twitter users is enough to achieve these levels of prediction and to know what will go viral across the entire Internet.
From Carlos III University of Madrid (Spain)
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found