The new system, which was developed in-house, learns which reviews are most helpful to customers and subsequently improves with time, according to Amazon spokeswoman Julie Law who was quoted by CNET as having said in an interview that the new system is intended “to make customer reviews more useful”.
Amazon is reportedly changing how it displays customer reviews on product pages in the U.S. It’s using machine learning to figure out which reviews are most helpful to users, and giving those more weight in the algorithm. Amazon gained much popularity with the reviews and the five-star rating system, helped buyers in deciding the popularity and quality of the products before purchasing.
Meanwhile, GeekWire also mentioned that Amazon is testing another system wherein customers can provide a numerical rating to a specific feature of an individual product. The improved system has already been introduced in the United States and Amazon has not yet confirmed whether the changes will be reflecting in other regions as well.
Amazon has taken the first steps to augment product reviews, one of the largest initiatives the company has taken in their 20 year history.
The adjustments in the algorithm will rank most recent reviews first. It could also work towards discrediting 1 star reviews purely left with the intention of dragging down a product’s score, possibly by competing vendors. But before handing over my cash, I made a decision to check out the reviews – and many weren’t so positive.
The new system will make note of the changes that a company makes to tweak or update its products, making them more noticeable to new customers. These include reviews that are purchased through fake review services, some of which Amazon has sued in the past.
The goal of course is to make the ratings and reviews on Amazon more reliable and up to date. Fake reviews and ratings generated by these sites threatened to tarnish the site’s reputation and destroy the trust of customers, sellers and manufacturers.
p style=”text-align: center;”>