Stanford researchers claim to have cracked an algorithm that will predict which photos have potential of going viral. “It wasn’t clear whether information cascades could be predicted because they happen so rarely,” said, assistant professor of computer science, Jure Leskovec. After a lot of indepth research of analysing anoymous data of thousands of Facebook users and bucketing the response and actions under various parameters the structure of sharing the photos was identified.
Based on this study these scientists noticed that they could accurately predict doubling the events almost 80 % of the time. As more photos got shared, their algorithm became even more accurate and stronger. Testimony to this were the photos that got shared 100s of times with their accuracy rate reached 88%.
LogicSpeak:
It has to be a wait and watch till tool is actually out for everyone to use. If it turns out to be as accurate then these scientists are going to be reach real soon as the marketers would pay any price to make their brands viral on a no-more-organic-friendly-platform like Facebook.
Link: https://bit.ly/1hwCQGs
