Published Paper Details:

A lot of people rely on content available on social media for making decisions. The possibility that anyone can post a review provides a golden opportunity for spammers to write spam reviews about products and services. Identifying these spammers and the spam content is a very important topic in field of research and although a considerable number of studies have been done recently, but so far, the methodologies put forth still barely detect spam reviews, and none of them show the importance of each extracted feature type. This propose a novel framework, named NetSpam, which utilizes spam features for modeling review datasets as heterogeneous information networks to map spam detection procedure into a classification problem in such networks. Using the importance of spam features help us to obtain better results in terms of different metrics experimented on real-world review datasets from Yelp and Amazon websites. The results show that NetSpam is better than the existing methods using the features like review-behavioral, user-behavioral, review-linguistic, user-linguistic.