Browse Prior Art Database

Method and System for Identifying Fake or Promotional Customer Reviews in Ecommerce Marketplaces

IP.com Disclosure Number: IPCOM000248313D
Publication Date: 2016-Nov-15
Document File: 3 page(s) / 187K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for identifying fake or promotional customer reviews in ecommerce marketplaces.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method and System for Identifying Fake or Promotional Customer Reviews in Ecommerce Marketplaces

Usually, ecommerce marketplaces rely on a reputation system to police sellers and provide a high quality experience for buyers. Since most of the ecommerce systems take a percentage of a product's price as a fee on each sale, sellers that are deemed to be of high quality with many sales are pushed to the top of the results in order to make more money form the ecommerce marketplace and provide quality products for the buyers. Being more visible to the visitors on the site has a direct relationship with how many sales a seller is able to make. The feedback loop is obvious to the sellers and provides sellers with a strong motivation to game the system. One of the ways to game the system as a seller on an ecommerce marketplace is to arrange a sale with a buyer and retrieves a stellar review and the highest rating on the product.

Ultimately, the result of gaming the system leads to poor shopping experience for the buyers, where the buyers cannot trust the reviews and rankings of the products any longer. The quality of the whole marketplace declines significantly, as sellers need to spend more time to go through the reviews and cannot rely on the star rating of the buyer as an indicator of quality.

Disclosed is a method and system for identifying fake or promotional customer reviews in ecommerce marketplaces.

In accordance with the method and system, sellers' data of newly created stores are retrieved from an ecommerce marketplace. Thereafter, the sellers' data such as, sell data etc., are filtered based on reviews and ratings. The retrieved reviews are analyzed for determining the sentiment of the reviews by using a sentiment analysis service. If the sentiment of the reviews is positive, the method and system extract the buyers data associated with the reviews. Thereafter, the method and system identifies and analyzes the reviews posted by each buyer on every store and categorizes the...