Unsupervised approach for detecting shilling attacks in collaborative recommender systems based on user rating behaviours. (1st May 2019)

• Record Type: Journal Article
• Authors: Zhang, Fuzhi; Ling, Zhoujun; Wang, Shilei
• Is Part Of: IET information security. Volume 13:Number 3(2019)
• Page Start: 174
• Page End: 187
• Publication Date: 2019-05-01
• Subjects: security of data -- recommender systems -- collaborative filtering; user preferences -- user rating item sequences -- unsupervised approach -- collaborative recommender systems -- detection performance -- use Gibbs latent Dirichlet allocation model -- shilling attacks detection -- Gibbs latent Dirichlet allocation model -- latent topic extraction; Computer security -- Periodicals; Cryptography -- Periodicals; Computer networks -- Security measures -- Periodicals; Database security -- Periodicals; 005.8
• DOI: 10.1049/iet-ifs.2018.5131
• Languages: English
• ISSNs: 1751-8709

Abstract

Abstract : Collaborative recommender systems have been known to be extremely vulnerable to shilling attacks. To prevent such attacks, many detection approaches including supervised and unsupervised have been proposed. However, the supervised approaches are only suitable for detecting known types of attacks and the unsupervised approaches require a priori knowledge to ensure the detection performance. To address the limitations, the authors propose an unsupervised approach for detecting shilling attacks based on user rating behaviours. They first use Gibbs latent Dirichlet allocation model to extract latent topics of user preferences from user rating item sequences, then they use mixture transition distribution model to construct the user's preference model and present several metrics to capture the diversity between genuine and attack users in rating behaviours. In the case of unknown attack size, the number of attack users is obtained by analysing the critical point of rating behaviour suspicious degrees between genuine and attack users, and based on which the attack users are identified. The experimental results on the MovieLens 1 M dataset show that the proposed approach outperforms the baseline methods in terms of recall and precision metrics.