A common vulnerability in online ecommerce sites is fraudulent manipulation of user profiles in order to boost the ratings of products in online recommender systems and overall reputation of the web site.
This article – Unsupervised Retrieval of Attack Profiles in Collaborative Recommender Systems casts this problem as a problem of detecting anomalous structure in network analysis and proposes a novel mechanism for detecting this anomalous structure.
Tell your friends and colleagues about us. Thanks!
Follow