Defending recommender systems: detection of profile injection attacks Export

@article{WMB07, abstract = {Abstract\ \ Collaborative recommender systems are known to be highly vulnerable to profile injection attacks, attacks that involve the insertion of biased profiles into the ratings database for the purpose of altering the system's recommendation behavior. Prior work has shown when profiles are reverse engineered to maximize influence; even a small number of malicious profiles can significantly bias the system. This paper describes a classification approach to the problem of detecting and responding to profile injection attacks. A number of attributes are identified that distinguish characteristics present in attack profiles in general, as well as an attribute generation approach for detecting profiles based on reverse engineered attack models. Three well-known classification algorithms are then used to demonstrate the combined benefit of these attributes and the impact the selection of classifier has with respect to improving the robustness of the recommender system. Our study demonstrates this technique significantly reduces the impact of the most powerful attack models previously studied, particularly when combined with a support vector machine classifier.}, author = {Williams, Chad and Mobasher, Bamshad and Burke, Robin}, citeulike-article-id = {3725456}, citeulike-linkout-0 = {http://dx.doi.org/10.1007/s11761-007-0013-0}, citeulike-linkout-1 = {http://www.springerlink.com/content/b82245815jk80165}, day = {30}, doi = {10.1007/s11761-007-0013-0}, journal = {Service Oriented Computing and Applications}, keywords = {attack, attacks, bias, collaborative, detection, filtering, injection, machines, models, profile, recommender, recommenders, securing, support, svm, systems, vector}, month = {November}, number = {3}, pages = {157--170}, posted-at = {2009-10-04 15:48:39}, priority = {0}, title = {Defending recommender systems: detection of profile injection attacks}, url = {http://dx.doi.org/10.1007/s11761-007-0013-0}, volume = {1}, year = {2007} }

TY - JOUR ID - WMB07 L3 - citeulike-article-id:3725456 N2 - Abstract  Collaborative recommender systems are known to be highly vulnerable to profile injection attacks, attacks that involve the insertion of biased profiles into the ratings database for the purpose of altering the system's recommendation behavior. Prior work has shown when profiles are reverse engineered to maximize influence; even a small number of malicious profiles can significantly bias the system. This paper describes a classification approach to the problem of detecting and responding to profile injection attacks. A number of attributes are identified that distinguish characteristics present in attack profiles in general, as well as an attribute generation approach for detecting profiles based on reverse engineered attack models. Three well-known classification algorithms are then used to demonstrate the combined benefit of these attributes and the impact the selection of classifier has with respect to improving the robustness of the recommender system. Our study demonstrates this technique significantly reduces the impact of the most powerful attack models previously studied, particularly when combined with a support vector machine classifier. IS - 3 JF - Service Oriented Computing and Applications EP - 170 TI - Defending recommender systems: detection of profile injection attacks VL - 1 SP - 157 KW - attack KW - attacks KW - bias KW - collaborative KW - detection KW - filtering KW - injection KW - machines KW - models KW - profile KW - recommender KW - recommenders KW - securing KW - support KW - svm KW - systems KW - vector AU - Williams, Chad AU - Mobasher, Bamshad AU - Burke, Robin PY - 2007/November/30/ UR - http://dx.doi.org/10.1007/s11761-007-0013-0 ER -