Sift Science Offers Faker Finder
October 15, 2012
There is no doubt that online fraud and fake reviews are a growing problem; Gigaom supplies “5 Ways to Sniff Out Online Fakers.” The article highlights the efforts of Sift Science, a new company co-founded by former Google software engineer Brandon Ballinger. In fact, six of the eight employees of the site-guarding startup are ex-Googlers.
“‘Based on user actions, we build a model of what a normal user would do on a site versus what a fraudulent user would do. We look at the time of account creation, the sequence of pages viewed. If they’re browsing around, they’re probably normal. If they set up an account and jump straight to a transaction, probably not,’ Ballinger told me by phone. But then again, they’re tricky. Sift Science found that someone who opens an account, then waits an hour before transacting is 7 times more likely to be fraudulent than the average user.
“The process is similar to Google Analytics in that Sift Science creates a history of user events and comes up with a score for each user that rates the likelihood that he or she is involved in fraud, he said.”
A beta trial of the software turned up some trends (the titular “5 Ways” to identify fraudsters). Fake accounts tend to be created late at night, for example. Also, these deceivers seem to use less current browsers and operating systems and really like Yahoo email. The geographic point of origin was also found to be significant.
We think Sift Science gives us a step in the right direction, but our question now is this: how can a search system differentiate valid research, verifiable facts, and “spin”? We’d love to see a few lines of code that can crack that one.
Cynthia Murrell, October 15, 2012