Detecting Spam Reviewers Based On Product Reviews

Recently, Evaluative texts on the Web have become a valuable source of opinions on products, services, events, individuals, etc. Recently, many researchers have studied such opinion sources as product reviews, forum posts, and blogs. However, existing research has been focused on classification and summarization of opinions using natural language processing and data mining techniques. An important issue that has been neglected so far is opinion spam or trustworthiness of online opinions. Hence, how to identify those spam reviews in a large number of reviews and mining spam reviewers is the precondition of sentiment analysis research, and then a big challenge.The Web has dramatically changed the way that people express themselves and interact with others. They can now post reviews of products at merchant sites and express their views and interact with others via blogs and forums. One example is shopping online, the review system has become a target of spammers who are usually hired or enticed by companies to write fake reviews to promote their products and services, and/or to distract customers from their competitors. Driven by profits, there are more and more spam reviews in major review websites.This paper focuses on spam reviews/reviewers detection on product or stores. We firstly make a conclusion of research status and other related works, then make a detail analysis of dataset and related algorithms, and then propose a novel multi-edges graph model to detect spammers. Our model constructed supportive edge set and conflicting edge set by reviewers opinion on products, then computing iteratively with reviewer features. Our algorithm is inspired by TrustRank but quite different with essential. At last, we run our algorithm on the multi-edges model, the evaluation result shows that our algorithm is effective and can reach a satisfied accuracy. In conclusion, our algorithm can detect the spam reviewers effectively, especially for those who always work together, which is the most harmful.