NET-SPAM: A Network Based Spam Detection Framework for Reviews in Online Social Media

Now a days, people confide on available content in social media in their decisions (e.g. reviews and feedback on a topic or product).For different interests and services, a spammers which can write spam reviews about their products that can leave a review. So far strategy used to detect spam reviews to show importance of each extracted feature type. A novel structure, named Net spam, which utilizes spam features for modeling review datasets as heterogeneous information networks to map spam detection procedure into classification problems in such networks. with the help of this features it help us to obtain better results for different experimented metrics on real-world review datasets from Amazon websites.Net Spam out performs the existing methods among four categories of features are; review-behavioral, user-behavioral, review linguistic, user-linguistic, review behavioral performs better than other categories.