A comparative study of shilling attack detectors for recommender systems

Author

Youquan Wang ; Lu Zhang ; Haicheng Tao ; Zhiang Wu ; Jie Cao

Author_Institution

Jiangsu Provincial Key Lab. of E-Bus., Nanjing Univ. of Finance & Econ., Nanjing, China

fYear

2015

fDate

22-24 June 2015

Firstpage

1

Lastpage

6

Abstract

Uncovering shilling attackers hidden in recommender systems is very crucial to enhance the robustness and trustworthiness of product recommendation. Many shilling attack detection algorithms have been proposed so far, and they exhibit complementary advantage and disadvantage towards various types of attacks. In this paper, we provide a thorough experimental comparison of several well-known detectors, including supervised and unsupervised methods. MovieLens 100K is the most widely-used dataset in the realm of shilling attack detection, and thus it is selected as the benchmark dataset. Meanwhile, seven types of shilling attacks generated by average-filling and random-filling model are tested in our experiments. As a result of our analysis, we show clearly causes and essential characteristics insider attackers that might determine the success or failure of different kinds of detectors.

Keywords

recommender systems; security of data; unsupervised learning; MovieLens 100K dataset; average-filling model; product recommendation; random-filling model; recommender systems; shilling attack detection; shilling attack detection algorithms; supervised methods; unsupervised methods; Covariance matrices; Detection algorithms; Detectors; Feature extraction; Niobium; Principal component analysis; Training; MovieLens; Recommender System; Shilling Attack Detection; Supervised Learning; Unsupervised Learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Service Systems and Service Management (ICSSSM), 2015 12th International Conference on

Conference_Location

Guangzhou

Print_ISBN

978-1-4799-8327-8

Type

conf

DOI

10.1109/ICSSSM.2015.7170330

Filename

7170330