Why Rumors spread fast in Social Networks

Information spread fast in social networks. This could be observed during recent demonstrations in the Arab world. Now computer scientists from the German Saarland University provide the mathematical proof for it and come up with a surprising explanation.

Tobias Friedrich

“It is fascinating”, Tobias Friedrich of the Cluster of Excellence on “Multimodal Computing and Interaction” says. He points out that so far you supposed that the uncontrolled growth in social networks creates a structure on which information spread very fast. “But now we can proof it in a mathematical way”, says Friedrich leading the independent research group “Random Structures and Algorithms”.

Together with his research colleagues Bejamin Doerr, adjunct professor for algorithms and complexity at Saarland University and researcher at Max Planck Institute for Informatics, and PhD student Mahmoud Fouz he proved that information spread in social networks much faster than in networks where everybody is communicating with everybody or in networks which structure is totally random based.

The scientists explain their results by the successful combination of persons with many contacts and persons with only a few contacts. “A person who keeps only a few connections can inform all its contacts very fast”, Friedrich says. Additionally you can prove that among these few contacts there always is a highly networked person who is contacted by a lot of other people in the social network, the scientist points out.  “Therefore everybody in this networks gets informed rapidly.”

To abstract the way of people connecting each other in a social network the scientists chose so called Preferential Attachment Graphs as basic network model. It assumes that new members of a social network rather connect to person maintaining many connections than to persons with only a few contacts. The communication within the network is based on the model that every person regularly exchanges all information with its contacts but never speaks to the  person it had contacted in the previous communication round.

It took the scientists twelve pages to write down the mathematical proof. They explain the concept of the proof easier to understand in the article “Why Rumors Spread Fast in Social Networks”, published in the magazine “Communications of the ACM” in June. It is peer reviewed.

Computer Science on the campus of Saarland University

A unique number of renowned computing science institutes do research on the campus in Saarbrucken, Germany. Additional to the computer science faculty and the Cluster on Excellence this are the German Research Center for Artificial Intelligence, the Max Planck Institute for Informatics, the Max Planck Institute for Software Systems, the Center for IT-Security, Privacy and Accountability and the Intel Visual Computing Institute.




newsarchive >>