Preventing shilling attacks in online recommender systems

330 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Poisoning Attacks to Graph-Based Recommender Systems (2018, background): “Formally, for a user u , RDMA is computed as follows: where I u is the set of items that user u has rated, | I u | is the number of items in I u , r ui is the rating score that u gave to item i , r i is the average rating score for item i , and o i is the total number of ratings for item i .”
Recommender Systems (2016, neutral): “In general, significant profit-driven motivations exist for some users to enter fake ratings [158, 329, 393, 444].”
Evaluating Recommendation Systems (2011, background): “empirically measuring average cost of a successful attack [36, 10].”
Comparison of collaborative filtering algorithms (2011, background): “—Shilling [Lam and Riedl 2004; Chirita et al. 2005].”
Recommender systems survey (2013, background): “RS can experience shilling attacks [128,57], which generate many positive ratings for a product, while products from competitors receive negative ratings.”
Shilling attacks against recommender systems: a comprehensive survey (2012, methodology): “(2004c) and Chirita et al. (2005) define cost of an attack as a function of the number of attack profiles added to the user-item matrix. In Tang and Tang (2011), Mobasher et al. (2005, 2007b), Williams (2006), Williams et al. (2007a), attack cost includes knowledge cost (necessary information about ”