Filters








5,832 Hits in 4.7 sec

Defending recommender systems: detection of profile injection attacks

Chad A. Williams, Bamshad Mobasher, Robin Burke
2007 Service Oriented Computing and Applications  
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  ...  This paper describes a classification approach to the problem of detecting and responding to profile injection attacks.  ...  To what extent does this detection ability succeed in defending the system against the influence of an attack?  ... 
doi:10.1007/s11761-007-0013-0 fatcat:vnomhfbzvbfo5lgyeimvpmmehu

Defending Grey Attacks by Exploiting Wavelet Analysis in Collaborative Filtering Recommender Systems [article]

Zhihai Yang
2015 arXiv   pre-print
"Shilling" attacks or "profile injection" attacks have always major challenges in collaborative filtering recommender systems (CFRSs).  ...  In this paper, we present a novel detection method to make recommender systems resistant to such attacks.  ...  presented in this paper is supported in part by the National Natural Science Foundation (61221063, U1301254), 863 High Tech Development Plan (2012AA011003) and 111 International Collaboration Program, of  ... 
arXiv:1506.05247v2 fatcat:vntyii3bj5ampp3pi3ssp4reqy

A genre trust model for defending shilling attacks in recommender systems

Li Yang, Xinxin Niu
2021 Complex & Intelligent Systems  
accuracy of system recommendations.  ...  Experimental results demonstrated the superior and comparable genre trust degrees recommended for defending against different types of shilling attacks.  ...  The detection of shilling attacks on recommender systems has been studied by many researchers.  ... 
doi:10.1007/s40747-021-00357-2 fatcat:r54hdarn3vfn7cbywbxuknadh4

Defending Grey Attacks by Exploiting Wavelet Analysis in Collaborative Filtering Recommender Systems

Zhihai, Zhongmin Cai, Agile Esmaeilikelishomi
2015 International Journal of Advanced Research in Artificial Intelligence (IJARAI)  
Shilling" attacks or "profile injection" attacks have always major challenges in collaborative filtering recommender systems (CFRSs).  ...  In this paper, we present a novel detection method to make recommender systems resistant to such attacks.  ...  between the number of items rated by user and the number of entire items in the recommender systems. where D is the set of the detected user profiles, is the set of attacker profiles, and is the set of  ... 
doi:10.14569/ijarai.2015.041103 fatcat:kvzfv7ltibeqhftm6gvr4ynahe

Defending Suspected Users by Utilizing Specific Distance Metric in Collaborative Filtering Recommender Systems

Bharat Vinod, Sonu, Phadtare, Gajanan Omkar, Siddhesh, Salunkhe
unpublished
Be that as it may, current CF strategies experience the ill effects of such issues as "shilling" attacks or "profile injection" attacks because of its openness.  ...  Recommender system is an imperative part of the data and internet business biological system. Collaborative filtering (CF) is a vital and well known innovation for recommender system.  ...  presented in this paper is supported in part by the National Natural Science Foundation (61221063, U1301254), 863 High Tech Development Plan (2012AA011003) and 111 International Collaboration Program, of  ... 
fatcat:fw5ee3zlwvfgxnsal3j7mrixiq

Toward trustworthy recommender systems

Bamshad Mobasher, Robin Burke, Runa Bhaumik, Chad Williams
2007 ACM Transactions on Internet Technology  
Recent research has begun to examine the vulnerabilities and robustness of different collaborative recommendation techniques in the face of "profile injection" attacks.  ...  Attackers, who cannot be readily distinguished from ordinary users, may inject biased profiles in an attempt to force a system to "adapt" in a manner advantageous to them.  ...  techniques for defending recommender systems against them.  ... 
doi:10.1145/1278366.1278372 fatcat:pv22pnijq5cetmz5bfrmxt7yjm

Detecting Abnormal Profiles in Collaborative Filtering Recommender Systems [article]

Zhihai Yang
2015 arXiv   pre-print
In this paper, we propose a novel detection method to make recommender systems resistant to the "shilling" attacks or "profile injection" attacks.  ...  The attackers can carefully inject chosen attack profiles into CFRSs in order to bias the recommendation results to their benefits.  ...  presented in this paper is supported in part by the National Natural Science Foundation (61221063, U1301254), 863 High Tech Development Plan (2012AA011003) and 111 International Collaboration Program, of  ... 
arXiv:1506.05752v3 fatcat:wl4ir2z5dfhu7i3rpi4hu5q54m

Attacking Recommender Systems with Augmented User Profiles [article]

Chen Lin, Si Chen, Hui Li, Yanghua Xiao, Lianyun Li, Qian Yang
2020 arXiv   pre-print
Recommendation Systems (RS) have become an essential part of many online services.  ...  In this paper we study the shilling attack: a subsistent and profitable attack where an adversarial party injects a number of user profiles to promote or demote a target item.  ...  On the contrary, most of the injected user profiles from conventional attack models can be easily detected.  ... 
arXiv:2005.08164v1 fatcat:auooejmsfzcatc2jpe6bxoyt5m

Classification features for attack detection in collaborative recommender systems

Robin Burke, Bamshad Mobasher, Chad Williams, Runa Bhaumik
2006 Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '06  
Attackers can use automated means to inject a large number of biased profiles into such a system, resulting in recommendations that favor or disfavor given items.  ...  Collaborative recommender systems are highly vulnerable to attack.  ...  ATTACK MODELS A profile injection attack against a recommender system consists of a set of attack profiles inserted into the system with the aim of altering the system's recommendation behavior with respect  ... 
doi:10.1145/1150402.1150465 dblp:conf/kdd/BurkeMWB06 fatcat:zkk4o2pumbfsbpqpit2uv3fwdm

Shilling Black-Box Recommender Systems by Learning to Generate Fake User Profiles

Chen Lin, Si Chen, Meifang Zeng, Sheng Zhang, Min Gao, Hui Li
2022 IEEE Transactions on Neural Networks and Learning Systems  
Conventional Shilling Attack approaches lack attack transferability (i.e., attacks are not effective on some victim RS models) and/or attack invisibility (i.e., injected profiles can be easily detected  ...  In this article, we study shilling attacks where an adversarial party injects a number of fake user profiles for improper purposes.  ...  Fig. 5 : 5 Fig. 5: Changes of Losses in Leg-UP. Fig. 6 : 6 Fig. 6: Attack detection of injected profiles (precision v.s. recall).  ... 
doi:10.1109/tnnls.2022.3183210 pmid:35749325 fatcat:2hfixbskkjgr7etrigsq4lxyfy

Defending Suspected Ratings in Collaborative Filtering Recommender Systems: A Fast Detection Method

Zhihai Yang
2016 Proceedings of The fourth International Conference on Information Science and Cloud Computing — PoS(ISCC2015)   unpublished
Collaborative filtering recommender systems (CFRSs) are key components of the well-known E-commerce websites such as Amazon, Yelp etc., to make personalized recommendations.  ...  We propose a fast and effective detection method to detect such attacks, which consists of two phases.  ...  However, CFRSs are highly vulnerable to shilling attacks due to their openness, which are injected with chosen profiles of abnormal ratings in order to control recommendation results to their benefits  ... 
doi:10.22323/1.264.0015 fatcat:mrjz5uyowbhcfobh6b7jigoe34

Preventing Recommendation Attack in Trust-Based Recommender Systems

Fu-Guo Zhang
2011 Journal of Computer Science and Technology  
We argue that trust-based recommender systems are facing novel recommendation attack which is different from the profile injection attacks in traditional recommender system.  ...  To the best of our knowledge, there has not any prior study on recommendation attack in a trust-based recommender system.  ...  We find that the method of directly injecting the biased profiles in a traditional recommender system is no longer valid in a trust-based recommender system, but there exists new recommendation attack  ... 
doi:10.1007/s11390-011-0181-4 fatcat:c4zxpokmu5gb3lffcxqvjb3ha4

Re-scale AdaBoost for attack detection in collaborative filtering recommender systems

Zhihai Yang, Lin Xu, Zhongmin Cai, Zongben Xu
2016 Knowledge-Based Systems  
Collaborative filtering recommender systems (CFRSs) are the key components of successful E-commerce systems. However, CFRSs are highly vulnerable to attacks due to its openness.  ...  Firstly, we extract well-designed features from user profiles based on the statistical properties of the diverse attack models, making hard detection scenarios become easier to perform.  ...  As a result, the overall quality of the paper has been noticeably enhanced, to which we feel much indebted and are grateful.  ... 
doi:10.1016/j.knosys.2016.02.008 fatcat:mvc3uh2vmjgkrjbc7xuhbhvq4m

Detecting shilling profiles in collaborative recommender systems via multidimensional profile temporal features

Yaojun Hao, Fuzhi Zhang
2018 IET Information Security  
To defend recommender systems, various methods have been proposed to detect shilling profiles, which can be categorised as user-and item-based detection methods.  ...  Most of the user-based methods identify shilling profiles via statistical signatures of rating values and suffer from low precision when detecting different types of attacks.  ...  The fake profiles injected are called shilling profiles, which have a negative impact on the recommendation results of recommender systems.  ... 
doi:10.1049/iet-ifs.2017.0012 fatcat:yc5qh2dkjnh4tcbfcvftnvlun4

A Survey on Trustworthy Recommender Systems [article]

Yingqiang Ge, Shuchang Liu, Zuohui Fu, Juntao Tan, Zelong Li, Shuyuan Xu, Yunqi Li, Yikun Xian, Yongfeng Zhang
2022 arXiv   pre-print
Recommender systems (RS), serving at the forefront of Human-centered AI, are widely deployed in almost every corner of the web and facilitate the human decision-making process.  ...  All of these create an urgent need for Trustworthy Recommender Systems (TRS) so as to mitigate or avoid such adverse impacts and risks.  ...  The size of an attack can be measured in several ways. The most frequently used measure is the number or percentage of profiles injected into the system by the attacker [294] .  ... 
arXiv:2207.12515v1 fatcat:lsnuwdtl5rboznmhhux2n5y5om
« Previous Showing results 1 — 15 out of 5,832 results