推荐系统中一种高效的攻击块挖掘算法

doi:10.3969/j.issn.1000-3428.2016.02.023

计算机工程

An Efficient Attack-block Mining Algorithm in Recommender System

QU Chunyan,GUAN Yating,LIU Guiquan

(School of Computer Science and Technology,University of Science and Technology of China,Hefei 230027,China)

Received:2015-02-12 Online:2016-02-15 Published:2016-01-29

摘要/Abstract

摘要： 针对现有在线推荐系统中协同过滤算法无法有效对抗文件注入攻击的问题,考虑目标攻击者和攻击项,定义攻击块的概念。结合攻击块中的块面积率(BAR)和块打分率(BRR)信息,提出一种高效的攻击块挖掘(MAB)算法,检测评分事务数据集上的攻击行为,并通过基于BAR与BRR 上界的剪枝策略,缩小攻击块的搜索空间及降低搜索耗时。在2个真实数据集上的实验结果表明,在不同攻击场景下MAB算法均能准确挖掘出攻击块,并且具有较高的挖掘效率。

关键词: 攻击块, 推荐系统, 块面积率, 块打分率, 块质量, 上界

Abstract: Aiming at the problem that the collaborative filtering algorithm in online recommender systems is very vulnerable to the profile injection attacks,this paper considers target attack item and attacker and defines the concept of attack-block.Combining with the information of Block Area Ratio(BAR) and Block Rate Ratio(BRR) contained in the attack-block respectively,this paper proposes an algorithm called Mining Attack-block(MAB) to detect the attack behavior on a given rating transaction database,and uses pruning strategy based on the upper bounds of BAR and BRR to narrow space and time for searching.Experimental results on two real data sets demonstrate the algorithm can detect attack-block in different attack scenarios,and it has higher efficiency than comparing algorithm.

Key words: attack-block, recommender system, Block Area Ratio(BAR), Block Rate Ratio(BRR), Block Quality(BQ), upper bound

中图分类号:

TP39

瞿春燕,管亚亭,刘贵全. 推荐系统中一种高效的攻击块挖掘算法[J]. 计算机工程, doi: 10.3969/j.issn.1000-3428.2016.02.023.

QU Chunyan,GUAN Yating,LIU Guiquan. An Efficient Attack-block Mining Algorithm in Recommender System[J]. Computer Engineering, doi: 10.3969/j.issn.1000-3428.2016.02.023.

http://www.ecice06.com/CN/Y2016/V42/I2/124

参考文献

参考文献［1］Jindal N,Liu Bing.Opinion Spam and Analysis［C］//Proceedings of International Conference on Web Search and Data Mining.New York,USA:ACM Press,2008:219-230. ［2］Li Fangtao,Huang Minglie,Yang Yi,et al.Learning to Identify Review Spam［C］//Proceedings of International Joint Conference on Artificial Intelligence.San Francisco,USA:AAAI Press,2011:2488-2492. ［3］Jinda N,Liu Bing,Lim E P.Finding Unusual Review Patterns Using Unexpected Rules［C］//Proceedings of International Conference on Information and Knowledge Management.New York,USA:ACM Press,2010:1549-1552. ［4］Mukherjee A,Liu Bing,Glance N.Spotting Fake Reviewer Groups in Consumer Reviews［C］//Proceedings of International Conference on World Wide Web.New York,USA:ACM Press,2012:191-200. ［5］Mukherjee A,Liu Bing,Wang J.Detecting Group Review Spam［C］//Proceedings of International Conference Companion on World Wide Web.New York,USA:ACM Press,2011:93-94. ［6］Su Xuefeng,Zeng Huajun,Chen Zheng.Finding Group Shilling in Recommendation System［C］//Proceedings of International Conference on World Wide Web.New York,USA:ACM Press,2005:960-961. ［7］Liu Yuhong,Sun Yan,Yu Ting.Defending Multiple-user-multiple-target Attacks in Online Reputation Systems［C］//Proceedings of the 3rd International Conference on Social Computing.Washington D.C.,USA:IEEE Press,2011:425-434. ［8］Koutrika G,Effendi F A,Gyongyi Z.Combating Spam in Tagging Systems［C］//Proceedings of International Workshop on Adversarial Information Retrieval on the Web.New York,USA:ACM Press,2007:57-64. ［9］Liu Jingjing,Cao Yunbo,Lin C Y,et al,Low-quality Product Review Detection in Opinion Summariza-tion［C］//Proceedings of Joint Conference on Empirical Methods in Natural Language Processing and Compute Natural Language Earning.Washington D.C.,USA:IEEE Press,2007:334-342. ［10］Burke R,Mobasher B,Williams C,et al.Classification Features for Attack Detection in Collaborative Recom-mender Systems［C］//Proceedings of International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2006:542-547. ［11］伍之昂,庄毅,王有权,等.基于特征选择的推荐系统托攻击检测算法［J］.电子学报,2012,40(8):1687-1693. ［12］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. ［13］Chirita P A,Nejdl W,Zamfir C.PreventingShilling Attacks in Online Recommender Systems［C］//Proceedings of International Workshop on Web Information and Data Management.New York,USA:ACM Press,2005:67-74. 编辑陆燕菲

[1]	唐彦, 卢镘旭. 基于知识图谱与深度涟漪网络的推荐系统[J]. 计算机工程, 2023, 49(5): 63-72,80.
[2]	李盼, 解庆, 李琳, 刘永坚. 知识增强的图神经网络序列推荐模型[J]. 计算机工程, 2023, 49(2): 70-80.
[3]	沈学利, 马玉营, 梁振兴. 融合复杂先验与注意力机制的变分自动编码器[J]. 计算机工程, 2022, 48(11): 55-61.
[4]	刘华玲, 马俊, 张国祥. 基于深度学习的内容推荐算法研究综述[J]. 计算机工程, 2021, 47(7): 1-12.
[5]	贾俊杰, 张玉超. 基于用户模糊聚类的综合信任推荐算法[J]. 计算机工程, 2021, 47(6): 60-67.
[6]	唐浩, 刘柏嵩, 刘晓玲, 黄伟明. 基于协同知识图谱特征学习的论文推荐方法[J]. 计算机工程, 2020, 46(9): 306-312.
[7]	吴清春, 贾彩燕. 一种融合社交关系的矩阵分解推荐模型[J]. 计算机工程, 2020, 46(8): 72-77,84.
[8]	李南星, 盛益强, 倪宏. 用于个性化推荐的条件卷积隐因子模型[J]. 计算机工程, 2020, 46(4): 85-90,96.
[9]	黄文明, 卫万成, 张健, 邓珍荣. 基于注意力机制与评论文本深度模型的推荐方法[J]. 计算机工程, 2019, 45(9): 176-182.
[10]	杨佳莉, 李直旭, 许佳捷, 赵朋朋, 赵雷, 周晓方. 一种自适应的混合协同过滤推荐算法[J]. 计算机工程, 2019, 45(7): 222-228.
[11]	毛德磊,唐雁. 基于归因理论用户偏好提取的协同过滤算法[J]. 计算机工程, 2019, 45(6): 225-229,236.
[12]	朱文跃,刘炜,刘宗田. 基于事件本体的新闻个性化推荐[J]. 计算机工程, 2019, 45(6): 267-272,279.
[13]	王泽华, 柯新生. 基于Coclus联合聚类与非负矩阵分解的推荐算法[J]. 计算机工程, 2019, 45(11): 68-73,80.
[14]	陈晓霞,卢菁. 融合多数据源的动态自适应推荐算法[J]. 计算机工程, 2018, 44(9): 64-69.
[15]	邢玉莹,夏鸿斌,王涵. 缺失数据建模的改进型ALS在线推荐算法[J]. 计算机工程, 2018, 44(8): 212-217,223.

选择文件类型/文献管理软件名称

选择包含的内容