《计算机应用研究》|Application Research of Computers

基于属性优化矩阵补全的抗托攻击推荐算法

Shilling-attack-tolerant recommendation algorithm based on attribute facilitated matrix completion

免费全文下载 (已被下载 次)  
获取PDF全文
作者 周宇轩,陈蕾,张涵峰
机构 南京邮电大学 a.计算机学院;b.江苏省无线传感网高技术研究重点实验室,南京 210003
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)03-017-0724-06
DOI 10.19734/j.issn.1001-3695.2017.09.0906
摘要 托攻击是当前推荐系统面临的严峻挑战之一。由于推荐系统的开放性,恶意用户可轻易对其注入精心设计的评分,从而影响推荐结果,降低用户体验。基于属性优化结构化噪声矩阵补全技术,提出一种鲁棒的抗托攻击个性化推荐(SATPR)算法。将攻击评分视为评分矩阵中的结构化行噪声,并采用L2,1范数进行噪声建模,同时引入用户与物品的属性特征以提高托攻击检测精度。实验表明,SATPR算法在托攻击下可取得比传统推荐算法更精确的个性化评分预测效果。
关键词 推荐系统;托攻击;L2,1范数正则化;属性特征
基金项目 江苏省自然科学基金面上项目(BK20161516)
中国博士后科学基金资助项目(2015M581794)
江苏省高校自然科学研究面上项目(15KJB520027)
江苏省博士后科研计划资助项目(1501023C)
本文URL http://www.arocmag.com/article/01-2019-03-017.html
英文标题 Shilling-attack-tolerant recommendation algorithm based on attribute facilitated matrix completion
作者英文名 Zhou Yuxuan, Chen Lei, Zhang Hanfeng
机构英文名 a.SchoolofComputerScience,b.JiangsuHighTechnologyResearchKeyLaboratoryforWirelessSensorNetworks,NanjingUniversityofPosts&Telecommunications,Nanjing210003,China
英文摘要 Shilling attack is one of serious challenges which recommender systems are facing. Malicious users can easily insert well-designed ratings into recommender systems to affect recommendation results and decrease user experiences because of the openness of recommender systems. This paper proposed a robust shilling-attack-tolerant personalized recommendation (SATPR) algorithm based on attribution facilitated matrix completion with structural noise technology, regarded the ratings of attack users in the rating matrix as structural row noise and modeled them with L2, 1-norm. This paper also introduced attributive cha-racters of users and items to improve the accuracy of detection of shilling-attack. Experimental results show that SATPR algorithm achieves more accurate results of personalized rating prediction than traditional recommendation algorithms under shilling attacks.
英文关键词 recommender system; shilling attack; L2,1-norm regularization; attributive characters
参考文献 查看稿件参考文献
  [1] Martin F J, Donaldson J, Ashenfelter A, et al. The big promise of recommender systems[J] . AI Magazine, 2011, 32(3):19-27.
[2] Koren Y, Bell R, Volinsky C. Matrix factorization techniques for re-commender systems[J] . Computer, 2009, 42(8):30-37.
[3] Gunes I, Kaleli C, Bilge A, et al. Shilling attacks against recommender systems:a comprehensive survey[J] . Artificial Intelligence Review, 2014, 42(4):767-799.
[4] Zhang Qiang, Luo Yuan, Weng Chuliang, et al. A trust-based detecting mechanism against profile injection attacks in recommender systems[C] //Proc of the 3rd IEEE International Conference on Secure Software Integration and Reliability Improvement. Washington DC:IEEE Computer Society, 2009:59-64.
[5] Bryan K, O’Mahony M P, Cunningham P. Unsupervised retrieval of attack profiles in collaborative recommender systems[C] //Proc of the 2nd ACM International Conference on Recommender System. New York:ACM Press, 2008:155-162.
[6] Li Wentao, Gao Min, Li Hua, et al. Shilling attack detection in re-commender systems via selecting patterns analysis[J] . IEICE Trans on Information & Systems, 2016, 99(10):2600-2611.
[7] Deng Zijun, Zhang Fei, Sandra P S, et al. Shilling attack detection in collaborative filtering recommender system by PCA detection and perturbation[C] //Proc of International Conference on Wavelet Analysis and Pattern Recognition. Piscataway, NJ:IEEE Press, 2016:213-218.
[8] 陈蕾, 杨庚, 陈正宇, 等. 基于结构化噪声矩阵补全的Web服务QoS预测[J] . 通信学报, 2015, 36(6):53-63. (Chen Lei, Yang Geng, Cheng Zhengyu, et al. Web services QoS prediction via matrix completion with structural noise[J] . Journal on Communications, 2015, 36(6):53-63. )
[9] Xiao Fu, Sha Chaoheng, Chen Lei, et al. Noise-tolerant localization from incomplete range measurements for wireless sensor networks[C] //Proc of IEEE Conference on Computer Communications. Piscataway, NJ:IEEE Press, 2015:2794-2802.
[10] Zhang Zhao, Li Fanzhang, Zhao Mingbo, et al. Robust neighborhood preserving projection by nuclear/L2, 1-norm regularization for image feature extraction[J] . IEEE Trans on Image Processing, 2017, 26(4):1607-1622.
[11] 汤镇宇, 孟凡荣, 王志晓. 基于稀疏表示的快速L2-范数人脸识别方法[J] . 计算机应用研究, 2016, 33(9):2831-2836. (Tang Zhen-yu, Meng Fanrong, Wang Zhixiao. Fast face recognition with regula-rized least square via sparse representation[J] . Application Research of Computers, 2016, 33(9):2831-2836. )
[12] Srebro N, Jaakkola T. Weighted low-rank approximations[C] //Proc of International Conference on Machine Learning. 2003:720-727.
[13] Candès E J, Recht B. Exact matrix completion via convex optimization[J] . Foundations of Computational Mathematics, 2009, 9(6):717-772.
[14] Cai Xiao, Nie Feiping, Huang Heng, et al. Multi-class L2, 1-norm support vector machine[C] //Proc of the 11th IEEE International Conference on Data Mining. Piscataway, NJ:IEEE Press, 2011:91-100.
[15] Xu Miao, Jin Rong, Zhou Zhihua. Speedup matrix completion with side information:application to multi-label learning[C] //Advances in Neural Information Processing Systems. 2013:2301-2309.
[16] Beck A, Tetruashvili L. On the convergence of block coordinate descent type methods[J] . SIAM Journal on Optimization, 2013, 23(4):2037-2060.
[17] Combettes P L, Wajs V R. Signal recovery by proximal forward-backward splitting[J] . SIAM Journal on Multiscale Modeling & Simulation, 2005, 4(4):1168-1200.
[18] Ng A Y, Jordan M I, Weiss Y. On spectral clustering:analysis and an algorithm[C] //Proc of International Conference on Neural Information Processing Systems:Natural and Synthetic. Cambridge, MA:MIT Press, 2001:849-856.
[19] Shi Chuan, Huang Yue, Yu P S, et al. HeteSim:a general framework for relevance measure in heterogeneous networks[J] . IEEE Trans on Knowledge and Data Engineering, 2014, 26(10):2479-2492.
[20] Shi Chuan, Liu Jian, Zhuang Fuzhen, et al. Integrating heterogeneous information via flexible regularization framework for recommendation[J] . Knowledge and Information Systems, 2016, 49(3):835-859.
[21] 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[J] . 自动化学报, 2011, 37(2):160-167. (Li Cong, Luo Zhigang, Shi Jinlong. An unsupervised algorithm for detecting shilling attacks on recommender systems[J] . Acta Automatica Sinica, 2011, 37(2):160-167. )
收稿日期 2017/9/4
修回日期 2017/11/20
页码 724-729,780
中图分类号 TP391;TP301.6
文献标志码 A