Communications of the ACM
Stanford Computer Scientists Learn to Predict Which Photos Will Go Viral on Facebook
Researchers at Stanford University have developed a method that allows them to predict whether a photo will go viral on Facebook.
Credit: Anna Cobb
Stanford University researchers have developed a method for predicting which photos on Facebook will go viral. Their method involves studying cascades, the term used to describe photos or videos being shared multiple times.
According to recent data provided by Facebook, only one in 20 photos posted on the social network gets shared even once, and just one in 4,000 gets more than 500 shares. The researchers were able to predict when a photo cascade would double in shares with 80-percent accuracy.
The researchers began by analyzing 150,000 Facebook photos, each of which had been shared at least five times. The researchers initially found that, at any given point in a cascade, there was a 50-percent chance the number of shares would double. The researchers then looked for variables that might help them predict doubling events more accurately than a coin toss, including the rate and speed at which photos were shared, as well as the structure of sharing. The algorithm became more accurate the more times a photo was shared, as photos shared hundreds of times had an accuracy rate of 88 percent.
From Stanford Report (CA)
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found