AN E-COMMERCE FEEDBACK REVIEW MINING FOR A TRUSTED SELLER'S PROFILE BY CLASSIFYING FAKE AND A UTHENTIC FEEDBACK COMMENTS

摘要

Before we make a purchase from an E-commerce site we usually browse through the reviews that are posted by the post purchase users. So reviews we find in an E-commerce site play a major role to help other user's in deciding whether to buy a product or not. Today lot of Reputation-based trust models are widely used in many E-commerce applications, and feedback ratings are computed to find sellers reputation trust scores. However the "all good reputation" problem is very common in E-commerce sites. Usually the reputation scores for sellers in an E-commerce site is very high and it is difficult for buyers to select trustworthy sellers. In this paper we consider users reviews in the form of text as well as reviews in the form of stars. The system design consists of five parts. They are (i) feedback comments Analysis, (ii) Mining of feedback comments, (iii) computation of dimensions weights and trust, (iv) classification of fake and authentic comments and v)seller trust profile.