A Framework for Detecting Spam Reviews in Online Social Media

Abstract

Now a days, reviews and feedback on Social media play a vital role in the aspect of shopping. Any person can drop their review or feedback in the reviews column and this provides a golden opportunity for the spammers. In this paper, we propose an efficient framework that identifies the spammers (spam reviews) by utilizing the spam features for modelling review datasets as Heterogenous Information Network to map spam detection procedure into a classification problem. Different datasets were collected from various sites like Amazon and Yelp and various experiments were conducted on them by utilizing the importance of spam features to obtain the best results.