University of California, Santa Barbara researchers, led by professor Ben Zhao, have developed machine-learning software that can identify crowdturfing, a term for falsifying one's popularity on social media sites.
The software uses 35 account characteristics, such as age and location, to recognize crowdturfers on China's version of Twitter with up to 99 percent accuracy.
Three years ago, Zhao and his team coined the term 'crowdturfing' when they showed that it constituted more than 80 percent of the activity on two leading crowdsourcing websites in China.
Meanwhile, Utah State University researchers led by professor Kyumin Lee also are studying crowdturfing, and recently published an analysis of Fiverr, a website in which people post "gigs" they are willing to do for $5. The Utah State researchers found 90 percent of the top 10 sellers on Fiverr were crowdturfing by selling Twitter followers, site traffic, or likes on Facebook.
The researchers developed software capable of detecting crowdturfing by analyzing key features of a gig with 97-percent accuracy.
From Technology Review
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA