基于node2vec的社交网络用户属性补全攻击

doi:10.3969/j.issn.1671-1122.2017.12.012

信息网络安全 ›› 2017, Vol. 17 ›› Issue (12): 67-72.doi: 10.3969/j.issn.1671-1122.2017.12.012

基于node2vec的社交网络用户属性补全攻击

裴杨^1,^2,³, 瞿学鑫^1,^2,³, 郭晓博¹, 段丁阳¹

1.中国科学院信息工程研究所,北京 100093
2.中国科学院数据与通信保护研究教育中心,北京 100093
3.中国科学院大学网络空间安全学院,北京 100049

收稿日期:2017-08-01 出版日期:2017-12-20 发布日期:2020-05-12
作者简介:
作者简介：裴杨（1994—）,男,河北,硕士研究生,主要研究方向为信息安全;瞿学鑫（1994—）,男,江西,硕士,主要研究方向为信息安全;郭晓博（1990—）,女,河北,研究实习员,硕士,主要研究方向为信息安全;段丁阳（1992—）,男,山西,研究实习员,硕士,主要研究方向为人工智能、知识图谱。
基金资助:
国家重点研发计划[2016YFB0800504]

User Attribute Completion Attack in Social Networks Based on Node2vec

Yang PEI^1,^2,³, Xuexin QU^1,^2,³, Xiaobo GUO¹, Dingyang DUAN¹

1.Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China
2.Center of Data Assurance and Communication Security, Chinese Academy of Sciences, Beijing 100093, China
3.School of Cyber Security, University of Chinese Academy of Sciences, Beijing 100049, China

Received:2017-08-01 Online:2017-12-20 Published:2020-05-12

摘要/Abstract

摘要：

针对社交网络内容安全,存在一种通过属性推断补全来获取用户私密属性的攻击。传统的基于无监督学习的方法和基于监督学习的属性补全攻击方法存在未能把结构相似性和同质性有效结合起来的问题。文章提出了一种基于隐式表达的用户属性补全攻击方法,把用户属性补全抽象为一个有监督的分类问题,基本思路是利用node2vec算法将社交网络中的用户节点映射成向量,然后将向量通过聚类方法计算一个节点所在的社区,在社区内构建分类模型,并利用此模型对用户缺失属性进行预测。文章在真实数据集上进行验证,证明了算法能够有效提高社交网络用户属性补全的准确率。

关键词: 属性补全, 同质性, 结构相似性, node2vec, 内容安全

Abstract:

In social networks, there is an attack threatening its content security by acquiring user private attributes from attribute inference completion. Traditional user attribute completion methods like unsupervised learning and supervised learning fail to effectively combine homogeneity with structural similarity. This paper presents a user attribute completion attacking method based on implicit expression, which abstracts user attribute completion as a supervised classification problem. The basic idea is to use node2vec algorithm to map the user nodes in social networks into vectors, and then use the clustering method to calculate the community where a node is located, construct the classification model in the community, and use this model to predict the missing attributes of the user. This paper verifies that this algorithm can improve the accuracy of user attribute completion in social networks on a real data set.

Key words: attribute completion, homogeneity, structural similarity, node2vec, content security

中图分类号:

TP309.1

裴杨, 瞿学鑫, 郭晓博, 段丁阳. 基于node2vec的社交网络用户属性补全攻击[J]. 信息网络安全, 2017, 17(12): 67-72.

Yang PEI, Xuexin QU, Xiaobo GUO, Dingyang DUAN. User Attribute Completion Attack in Social Networks Based on Node2vec[J]. Netinfo Security, 2017, 17(12): 67-72.

图/表 6

参考文献 19

[1]	秦锐. 中国社交网络的兴起与发展[J]. 赤峰学院学报:自然科学版,2016(17):267-268.
[2]	温俊伟. 面向Twitter的分析系统研究[J]. 信息网络安全,2016(9):234-239.
[3]	冯泽奇. 大数据时代网络隐私问题研究:从权利保护到个人选择[D].长春:吉林大学,2016.
[4]	刘奇飞. 基于兴趣的微博用户关系分析原型系统研究[J]. 信息网络安全,2016(9):240-245.
[5]	RAO D,YAROWSKY D.Classifying Latent User Attributes in Twitter[C]//ACM.Association for Computing Machinery,October 30 , 2010,Toronto,Canada .New York: ACM,2010:37-44.
[6]	MISLOVE A,VISWANATH B,GUMMANDI K P.You Are Who You Know : Inferring User Profiles in Online Social Networks[C]// ACM.Association for Computing Machinery,February 4 - 6, 2010, New York, USA. New York: ACM ,2010:251-260.
[7]	COSCIA M,ROSSETI G,GIANNOTTI F.DEMON: A Localfirst Discovery Method for Overlapping Communities[C]// ACM .18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,August 12-16,Beijing,China. New York: ACM,2012:615–623.
[8]	PETER D H, ADRIAN E R, MARK. S H.Latent Space Approaches to Social Network Analysis[J].American Statistical Association, 2002,97(460):1090-1098.
[9]	GROVER A, LESKOVEC J.Node2vec: Scalable Feature Learning for Networks[C]// ACM.Association for Computing Machinery,ACM SIGKDD International Conference,August 13-17, 2016, San Francisco, California ,USA.New York:ACM, 2016:855-864.
[10]	李乐, 章毓晋. 非负矩阵分解算法综述[J]. 电子学报, 2008, 36(4):737-743.
[11]	EGGERT J, KORNER E.Sparse Coding and NMF[C]// IEEE .IEEE International Joint Conference on Neural Networks, 2004,July 25-29 ,2004 ,Budapest, Hungary . New York:IEEE, 2004:2529-2533.
[12]	JIA Y, ZHANG C, LU Q.Users' Brands Preference Based on SVD++ in Recommender Systems[C]// IEEE .IEEE Workshop on Advanced Research and Technology in Industry Applications,September 29-30 ,2014 ,Ottawa, ON, Canada . New York:IEEE, 2014:1175-1178.
[13]	CAO J, HU H, LUO T. Distributed Design and Implementation of SVD++ Algorithm for E-commerce Personalized Recommender System[EB/OL]. .
[14]	KUMAR R, KUMAR V B, SUNDER R S.Social Popularity Based SVD++ Recommender System[J]. International Journal of Computer Applications, 2014, 87(14):33-37.
[15]	LIAO L, HO Q, JIANG J.SLR: A Scalable Latent Role Model for Attribute Completion and Tie Prediction in Social Networks[C]// IEEE, International Conference on Data Engineering,May 16-20, 2016,Helsinki, Finland. New York:IEEE, 2016:1062-1073.
[16]	YIN X,WU B,LIN X.A Unified Framework for Predicting Attributes and Links in Social Networks[C]//IEEE International Conference on Big Data, December 23,2013,Silicon Valley, CA, USA. New York: IEEE,2013:153-160.
[17]	丁宇新,肖骁,吴美晶,等. 基于半监督学习的社交网络用户属性预测[J]. 通信学报,2014,(8):15-22.
[18]	王锐,吴玲玲,石川,等. 基于社团结构的链接预测和属性推断联合解决方法[J]. 电子学报,2016,(9):2062-2067.
[19]	徐丽萍,郝文江. 美国政企网络威胁情报现状及对我国的启示[J]. 信息网络安全,2016(9):278-284.

基于node2vec的社交网络用户属性补全攻击

User Attribute Completion Attack in Social Networks Based on Node2vec

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 19

相关文章 9

编辑推荐

Metrics

本文评价

[1]	车翔玖, 胡天岳. 基于node2vec神经网络的信息取证方案研究[J]. 信息网络安全, 2018, 18(10): 24-30.
[2]	崔鹏飞, 裘玥, 孙瑞. 面向网络内容安全的图像识别技术研究[J]. 信息网络安全, 2015, 15(9): 154-157.
[3]	李娇;王健;曹继文;陈彤. 基于感知哈希的在线发表论文版权保护系统[J]. , 2013, 13(11): 0-0.
[4]	张华玲;陈小军. 网络数据流内容安全检测算法优化与实现[J]. , 2013, 13(10): 0-0.
[5]	王庆广;何力;韩伟红. 基于爬虫的有害网站发现与判别系统的实现[J]. , 2012, 12(8): 0-0.
[6]	封莎;魏园园. 三网融合安全问题分析[J]. , 2010, (2): 0-0.
[7]	李留英. 各国互联网内容安全监管现状比较与分析[J]. , 2010, (1): 0-0.
[8]	江涛;于洪志;徐涛. 互联网藏文内容安全检测过滤系统研究[J]. , 2009, 9(10): 0-0.
[9]	董皓;张楚. 信息网络安全的法学定义研究--从技术视角向法律思维的转换[J]. , 2006, 6(2): 0-0.