The website at www.fakespot.com/ was created to “Bring trust back to the Internet” say the owners.
Fakespot is a data analytics company that wants to change the way people read reviews and similar content.
They believe that authentic user reviews are just about the best thing to come out of the Internet. However, the user review system is often abused by sellers that pay for reviews, by companies trying to make their competition look bad, and technologies that pretend to be real reviewers.
How to Use Fakespot
Fakespot can scan all of the reviews for any product or service you select on Yelp™ or Amazon™, Trip Advisor and Apple APP store and tell you whether the reviews are generally reliable or generally unreliable through the letter grade system.
With so many online shopping options, a strong or weak product review can have a huge impact on whether or not a purchase is made. The credibility of these reviews is undermined by businesses who leave fake reviews for themselves or for their competitors – or by individuals with an undisclosed bias.
Fakespot does not review products so cannot tell you how good a product is, it simply analyses the existing reviews looking for patterns that indicate authenticity or otherwise.
Fakespot uses various techniques to evaluate the authenticity of reviews, including:-.
- English language pattern recognition
- The profile of the reviewer
- Correlation with other reviewer data
The algorithm uses machine learning to constantly improve itself by looking at profile clusters, sentiment analysis and cluster correlation. We use artificial intelligence that has been trained to pick up on patterns. The more data that flows into the system, the better the system gets at the detecting fakes.
Amazon unverified reviews are considered unreliable by Fakespot because when the system associates a product review with a product purchase, that review is from a “verified purchaser”. These reviews are in most cases reliable, since Amazon has already confirmed an actual purchase of the product being reviewed.
But, if an Amazon review is not from a ‘verified purchaser’ there is no way of knowing for sure if the reviewer even used the product. While it is possible that a reviewer could have purchased the product elsewhere and left a review on Amazon at a later date, without purchase verification, it is impossible to tell.
Also, Fakespot systems have shown that most paid reviews come from unverified purchasers.
Do leave a comment on this post – click on the post title then scroll down to leave your comment.