Semi-SAD: applying semi-supervised learning to shilling attack detection

Authors:
Zhiang Wu;Jie Cao;Bo Mao;Youquan Wang
Affiliations:
Nanjing University of Finance and Economics, Nanjing, China;Nanjing University of Finance and Economics, Nanjing, China;Royal Institute of Technology, Stockholm, Sweden;Nanjing University of Finance and Economics, Nanjing, China
Venue:
Proceedings of the fifth ACM conference on Recommender systems
Year:
2011

Citing 5
Cited 2

Text Classification from Labeled and Unlabeled Documents using EM

Machine Learning - Special issue on information retrieval
Classification features for attack detection in collaborative recommender systems

Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
Unsupervised strategies for shilling detection and robust collaborative filtering

User Modeling and User-Adapted Interaction
Statistical attack detection

Proceedings of the third ACM conference on Recommender systems
Analysis and detection of segment-focused attacks against collaborative recommendation

WebKDD'05 Proceedings of the 7th international conference on Knowledge Discovery on the Web: advances in Web Mining and Web Usage Analysis

Review quality aware collaborative filtering

Proceedings of the sixth ACM conference on Recommender systems
Shilling attack detection utilizing semi-supervised learning method for collaborative recommender system

World Wide Web

Quantified Score

Hi-index	0.00

Visualization

Abstract

Collaborative filtering (CF) based recommender systems are vulnerable to shilling attacks. In some leading e-commerce sites, there exists a large number of unlabeled users, and it is expensive to obtain their identities. Existing research efforts on shilling attack detection fail to exploit these unlabeled users. In this article, Semi-SAD, a new semi-supervised learning based shilling attack detection algorithm is proposed. Semi-SAD is trained with the labeled and unlabeled user profiles using the combination of naïve Bayes classifier and EM-», augmented Expectation Maximization (EM). Experiments on MovieLens datasets show that our proposed Semi-SAD is efficient and effective.