A growing problem in the world of online reviewing is the prevalence of false reviews, or deceptive opinion spam, which are "fictitious opinions that have been deliberately written to sound authentic, in order to deceive the reader," explains a recent Cornell University paper [pdf].
The New York Times recently explored the phenomenon of help-for-hire sites that will pay people to post positive reviews, regardless of the restaurant's actual quality -- or even without ever setting foot inside -- of certain establishments or products. As fake reviews amass, the proliferation of misinformation grows and sites begin to lose their credibility as unbiased review forums. Which perhaps explains why after the Cornell team published the paper, in which they developed a methodology to ferret out fake reviews, the researchers were contacted by over a dozen companies to help them spot the fake reviews.
Beyond pay-for-play positive reviews, there are also documented instances of downright inaccuracies that lead to false, but negative, reviews. Graham Elliot's Grahamwich in Chicago started getting negative reviews before it even opened. More recently, Brooklyn restaurant Anella requested that a fake review get removed, and Yelp declined.
Then again, there's also the problem of not necessarily fake reviews, but the ultimately useless ones, as is chronicled on the Tumblr Fuck You Yelper. Perhaps the Cornell researchers can develop an algorithm to filter out idiocy as well.