一种基于流行度分类特征的托攻击检测算法

李文涛; 高旻; 李华; 熊庆宇; 文俊浩; 凌斌

doi:10.16383/j.aas.2015.c150040

一种基于流行度分类特征的托攻击检测算法

doi: 10.16383/j.aas.2015.c150040

李文涛^1,2,
高旻^1,3, ,,
李华^1,2,
熊庆宇^1,3,
文俊浩^1,3,
凌斌⁴

1.
信息物理社会可信服务计算教育部重点实验室重庆 400044, 中国;
2.
重庆大学计算机学院重庆 400044, 中国;
3.
重庆大学软件学院重庆 400044, 中国;
4.
英国朴茨茅次大学工程学院朴茨茅次 PO1 3AH, 英国

基金项目:

国家重点基础研究发展计划(973计划)(2013CB328903),国家自然科学基金(71102065),重庆市基础与前沿研究计划项目(cstc2015jcyjA40049),中国博士后基金(2012M521680),中央高校基础研究基金(106112014CDJZR095502,CDJZR12090001)资助

详细信息

作者简介:
李文涛重庆大学计算机学院硕士研究生.主要研究方向为个性化推荐与数据挖掘.E-mail:livent@126.com

李华重庆大学计算机学院副教授.主要研究方向为计算机网络,数据挖掘与大数据.E-mail:LH@cqu.edu.cn

熊庆宇重庆大学软件学院教授.主要研究方向为人工神经网络,量子神经计算及其应用.E-mail:xiong03@cqu.edu.cn

文俊浩重庆大学软件学院教授.主要研究方向为计算智能及服务计算.E-mail:jhwen@cqu.edu.cn

凌斌英国朴茨茅次大学电子工程学院研究员.主要研究方向为信息共享,项目管理,推荐系统.E-mail:bin.ling@myport.ac.uk

通讯作者:
高旻重庆大学软件学院副教授.主要研究方向为个性化推荐,服务计算,数据挖掘.本文通信作者.E-mail:mingaoo@gmail.com

计量
- 文章访问数: 2664
- HTML全文浏览量: 131
- PDF下载量: 1242
- 被引次数: 0
出版历程
- 收稿日期: 2015-01-26
- 修回日期: 2015-06-01
- 刊出日期: 2015-09-20

An Shilling Attack Detection Algorithm Based on Popularity Degree Features

LI Wen-Tao^1,2,
GAO Min^{1,3
, ,},
LI Hua^1,2,
XIONG Qing-Yu^1,3,
WEN Jun-Hao^1,3,
LING Bin⁴

1.
Key Laboratory of Dependable Service Computing in Cyber Physical Society, Ministry of Education Chongqing 400044, China;
2.
College of Computer Science, Chongqing University, Chongqing 400044, China;
3.
School of Software Engineering, Chongqing University, Chongqing 400044, China;
4.
School of Engineering, University of Portsmouth, Portsmouth PO1 3AH, UK

Funds:

Supported by National Key Basic Research Program of China (973 Program) (2013CB328903), National Natural Science Foundation of China (71102065), Basic and advanced research projects in Chongqing (cstc2015jcyjA40049), China Postdoctoral Science Foundation (2012M521680), and Fundamental Research Funds for the Central Universities (106112014CDJZR095502, CDJZR12090001)

摘要

摘要: 基于协同过滤的推荐系统容易受到托攻击的危害, 如何检测托攻击成为推荐系统可靠性的关键. 针对现有托攻击检测手段使用基于评分的分类特征易受混淆技术干扰的局限, 本文从用户选择评分项目方式入手, 分析由此造成的用户概貌中已评分项目的流行度分布情况的不同, 提出用于区分正常用户与虚假用户基于流行度的分类特征, 进而得到基于流行度的托攻击检测算法. 实验表明该算法在托攻击检测中具有更强的检测性能与抗干扰性.
- 协同过滤 /
- 托攻击 /
- 项目流行度 /
- 幂律分布 /
- 基于流行度的分类特征
Abstract: Recommendation systems based on collaborative filtering are vulnerable to shilling attacks, so how to detect attacks becomes crucial to ensure the reliability of these systems. Because the current shilling attack detection methods based on features extracted from rating patterns are susceptible to obfuscation technologies, this paper starts from a statistics analysis of the way users choose items to rate, thus getting the corresponding results of different rated items popularity degree (rated times) distributions in normal users's profiles and spam users' profile. Then classification features based on popularity degree are proposed to distinguish these two types of users. Finally, a shilling attack detection algorithm based on popularity features is developed. Experiments show that the detection performance of the algorithm is superior in attack detection precision and interference resistance.
- Collaborative filtering /
- shilling attack /
- item popularity /
- power-law distribution /
- features based on popularity

HTML全文

参考文献(27)

[1]	Borrás J, Moreno A, Valls A. Intelligent tourism recommender systems: a survey. Expert Systems with Applications, 2014, 41(16): 7370-7389
[2]	Qu M, Zhu H S, Liu J M, Liu G N, Xiong H. A cost-effective recommender system for taxi drivers. In: Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2014. 45-54
[3]	Chung N, Koo C, Kim J K. Extrinsic and intrinsic motivation for using a booth recommender system service on exhibition attendees' unplanned visit behavior. Computers in Human Behavior, 2014, 30: 59-68
[4]	Gao M, Wu Z F, Jiang F. Userrank for item-based collaborative filtering recommendation. Information Processing Letters, 2011, 111(9): 440-446
[5]	Li Cong, Luo Zhi-Gang. A metadata-enhanced variational bayesian matrix factorization model for robust collaborative recommendation. Acta Automatica Sinica, 2011, 37(9): 1067-1076(李聪, 骆志刚. 用于鲁棒协同推荐的元信息增强变分贝叶斯矩阵分解模型. 自动化学报, 2011, 37(9): 1067-1076)
[6]	Li C, Luo Z G. Detection of shilling attacks in collaborative filtering recommender systems. In: Proceedings of the 2011 International Conference of Soft Computing and Pattern Recognition (SoCPaR). Dalian, China: IEEE, 2011. 190-193
[7]	Mobasher B, Burke R, Williams C, Bhaumik R. Analysis and detection of segment-focused attacks against collaborative recommendation. In: Proceeding of the 7th International Workshop on Knowledge Discovery on the Web, Advances in Web Mining and Web Usage Analysis. Berlin, Heidelberg: Springer, 2006. 96-118
[8]	Seminario C E, Wilson D C. Attacking item-based recommender systems with power items. In: Proceedings of the 8th ACM Conference on Recommender Systems. New York: ACM, 2014. 57-64
[9]	Xia H, Fang B, Gao M, Ma H, Tang Y Y, Wen J. A novel item anomaly detection approach against shilling attacks in collaborative recommendation systems using the dynamic time interval segmentation technique. Information Sciences, 2015, 306: 150-165
[10]	Zhang Z, Kulkarni S R. Detection of shilling attacks in recommender systems via spectral clustering. In: Proceedings of the 17th International Conference on Information Fusion (FUSION). Salamanca: IEEE, 2014. 1-8
[11]	Gunes I, Kaleli C, Bilge A, Polat H. Shilling attacks against recommender systems: a comprehensive survey. Artificial Intelligence Review, 2014, 42(4): 767-799
[12]	Wu Zhi-Ang, Wang You-Quan, Cao Jie. A survey on shilling attack models and detection techniques for recommender systems. Chinese Science Bulletin, 2014, 59(7): 551-560(伍之昂, 王有权, 曹杰. 推荐系统托攻击模型与检测技术. 科学通报, 2014, 59(7): 551-560)
[13]	Wu Zhi-Ang, Zhuang Yi, Wang You-Quan, Cao Jie. Shilling attack detection based on feature selection for recommendation systems. Acta Electronica Sinica, 2012, 40(8): 1687-1693(伍之昂, 庄毅, 王有权, 曹杰. 基于特征选择的推荐系统托攻击检测算法. 电子学报, 2012, 40(8): 1687-1693)
[14]	Lam S K, Riedl J. Shilling recommender systems for fun and profit. In: Proceedings of the 13th International Conference on World Wide Web. New York: ACM, 2004. 393-402
[15]	Oestreicher-Singer G, Sundararajan A. Recommendation networks and the long tail of electronic commerce. MIS Quarterly, 2012, 36(1): 65-84
[16]	Yin H Z, Cui B, Li J, Yao J J, Chen C. Challenging the long tail recommendation. Proceedings of the VLDB Endowment, 2012, 5(9): 896-907
[17]	Li Cong, Luo Zhi-Gang. Detecting shilling attacks in recommender systems based on non-random-missing mechanism. Acta Automatica Sinica, 2013, 39(10): 1681-1690(李聪, 骆志刚. 基于数据非随机缺失机制的推荐系统托攻击探测. 自动化学报, 2013, 39(10): 1681-1690)
[18]	Li Cong, Luo Zhi-Gang, Shi Jin-Long. An unsupervised algorithm for detecting shilling attacks on recommender systems. Acta Automatica Sinica, 2011, 37(2): 160-167(李聪, 骆志刚, 石金龙. 一种探测推荐系统托攻击的无监督算法. 自动化学报, 2011, 37(2): 160-167)
[19]	Chirita P A, Nejdl W, Zamfir C. Preventing shilling attacks in online recommender systems. In: Proceedings of the 7th Annual ACM International Workshop on Web Information and Data Management. New York: ACM, 2005. 67-74
[20]	Burke R, Mobasher B, Williams C, Bhaumik R. Classification features for attack detection in collaborative recommender systems. In: Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2006. 542-547
[21]	Mehta B, Nejdl W. Unsupervised strategies for shilling detection and robust collaborative filtering. User Modeling and User-Adapted Interaction, 2009, 19(1-2): 65-97
[22]	Wu Z, Wu J J, Cao J, Tao D C. HySAD: a semi-supervised hybrid shilling attack detector for trustworthy product recommendation. In: Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2012. 985-993
[23]	Zipf G K. Selected studies of the principle of relative frequency in language. Language, 1932, 9(1): 89-92
[24]	Zhou Tao, Han Xiao-Pu, Yan Xiao-Yong, Yang Zi-Mo, Zhao Zhi-Dan, Wang Bing-Hong. Statistical mechanics on temporal and spatial activities of human. Journal of University of Electronic Science and Technology of China, 2013, 42(4): 481-540(周涛, 韩筱璞, 闫小勇, 杨紫陌, 赵志丹, 汪秉宏. 人类行为时空特性的统计力学. 电子科技大学学报, 2013, 42(4): 481-540)
[25]	Lv L, Zhang Z K, Zhou T. Zipf's law leads to heaps' law: analyzing their relation in finite-size systems. PloS One, 2010, 5(12): e14139
[26]	Mobasher B, Burke R, Bhaumik R, Sandvig J J. Attacks and remedies in collaborative recommendation. IEEE Intelligent Systems, 2007, 22(3): 56-63
[27]	Chen J, Luo D L, Mu F X. An improved ID3 decision tree algorithm. In: Proceedings of the 4th International Conference on Computer Science & Education. Nanning: IEEE, 2009. 127-130