在线社交网络中Spam相册检测方案

doi:10.11959/j.issn.1000-436x.2016180

通信学报 ›› 2016, Vol. 37 ›› Issue (9): 75-91.doi: 10.11959/j.issn.1000-436x.2016180

在线社交网络中Spam相册检测方案

吕少卿¹,张玉清^1,²,刘东航¹,张光华^1,³

¹ 西安电子科技大学综合业务网理论及关键技术国家重点实验室，陕西西安710071
² 中国科学院大学国家计算机网络入侵防范中心，北京 100190
³ 中国科学院信息工程研究所物联网信息安全技术北京市重点实验室,北京 100097

出版日期:2016-09-25 发布日期:2016-09-28
基金资助:
国家自然科学基金资助项目;国家自然科学基金资助项目;国家自然科学基金资助项目;物联网信息安全技术北京市重点实验室开放课题基金资助项目;中国博士后科学基金资助项目

Detecting Spam albums in online social network

Shao-qing LYU¹,Yu-qing ZHANG^1,²,Dong-hang LIU¹,Guang-hua ZHANG^1,³

¹ Information Security Research Center of State Key Laboratory of Integrated Services Networks,Xidian University,Xi'an 710071,China
² National Computer Network Intrusion Protection Center,University of Chinese Academy of Sciences,Beijing 100190,China
³ Beijing Key Laboratory of IOT Information Security Technology,Institute of Information Engineering,CAS,Beijing 100097,China

Online:2016-09-25 Published:2016-09-28
Supported by:
The National Natural Science Foundation of China;The National Natural Science Foundation of China;The National Natural Science Foundation of China;Open Fund of Beijing Key Laboratory of IOT Information Security Technology;China Postdoctoral Science Foundation

摘要/Abstract

摘要：

提出一种针对Spam相册的检测方案。首先分析了Photo Spam的攻击特点以及与传统Spam的差异，在此基础上构造了12个提取及时且计算高效的特征。利用这些特征提出了有监督学习的检测模型，通过2 356个相册的训练形成Spam相册分类器，实验表明能够正确检测到测试集中100%的Spam相册和98.2%的正常相册。最后将训练后的模型应用到包含315 115个相册的真实数据集中，检测到89 163个Spam相册，正确率达到97.2%。

关键词: 社交网络安全, PhotoSpam, Spam检测, 人人网

Abstract:

A supervised learning solution to detect Spam albums instead of spammers in Photo Spam was proposed.Specifically,the characteristics of Photo Spam and the differences between Photo Spam and traditional Spam were analyzed.Then 12 features which were extracted easily and calculated efficiently were constructed based on the analysis.Next a classification model was built with a dataset of 2 356 labeled albums to identify Spam albums.The model provided excellent performance with true positive rates of Spam albums and normal albums,reaching 100% and 98.2% respectively.Finally,the detection model were applied to 315 115 unlabeled albums and detected 89 163 spam albums with a true positive rate of 97.2%.

Key words: social network security, Photo Spam, Spam detection, RenRen

吕少卿,张玉清,刘东航,张光华. 在线社交网络中Spam相册检测方案[J]. 通信学报, 2016, 37(9): 75-91.

Shao-qing LYU,Yu-qing ZHANG,Dong-hang LIU,Guang-hua ZHANG. Detecting Spam albums in online social network[J]. Journal on Communications, 2016, 37(9): 75-91.

图/表 13

表1

图1

图2

图3

表2

图4

表3

表4

表5

表6

图5

表7

表8

参考文献 27

[1]	GAO H , HU J , HUANG T . Security issues in online social networks[J]. IEEE Internet Computing, 2011,15(4): 56-63.
[2]	FIRE M , GOLDSCHMIDT R , ELOVICI Y . Online social networks:threats and solutions survey[J]. IEEE Communications Surveys and Tutorials, 2013,16(4): 2019-2036.
[3]	CAVIGLIONE L , COCCOLI M , MERLO A . A taxonomy–based model of security and privacy in online social networks[J]. International Journal of Computational Science and Engineering, 2014,9(4): 325-338.
[4]	GAO H , HU J , WILSON C . Detecting and characterizing social Spam campaigns[C]// The 10th ACM SIGCOMM conference on Internet Measurement. Melbourne,Australia, 2010: 35-47.
[5]	LEE K , EOFF B D , CAVERLEE J . Seven months with the devils:a long-term study of content polluters on twitter[C]// The Fifth International AAAI Conference on Weblogs and Social Media. Barcelona,Spain, 2011: 185-192.
[6]	MARTINEZ-ROMO J , ARAUJO L . Detecting malicious tweets in trending topics using a statistical analysis of language[J]. Expert Systems with Applications, 2013,40(8): 2992-3000.
[7]	MILLER Z , DICKINSON B , DEITRICK W . Twitter Spammer detection using data stream clustering[J]. Information Sciences, 2014,260: 64-73.
[8]	ZHANG J , GU G . NEIGHBORWATCHER:A content-agnostic comment spam inference system[C]// Network & Distributed System Security Symposium. San Diego,CA,United States, 2013.
[9]	BENEVENUTO F , MAGNO G , RODRIGUES T . Detecting spammers on twitter[C]// The Collaboration,Electronic Messaging,Anti-abuse and Spam Conference. Redmond,USA, 2010: 6-12.
[10]	ZHENG X , ZENG Z , CHEN Z . Detecting spammers on social networks[J]. Neurocomputing, 2015,159(2): 27-34.
[11]	STRINGHINI G , KRUEGEL C , VIGNA G . Detecting spammers on social networks[C]// The 26th Annual Computer Security Applications Conference. TX,USA, 2010: 1-9.
[12]	LEE K , CAVERLEE J , WEBB S . Uncovering social Spammers:social honeypots+ machine learning[C]// The 33rd International ACM SIGIR Conference on Research and Development in Information Retrieval. Geneva,Switzerland, 2010: 435-442.
[13]	ZHU Y , WANG X , ZHONG E . Discovering Spammers in social networks[C]// The 26th AAAI Conference on Artificial Intelligence. Toronto,Canada, 2012: 171-177.
[14]	WANG G , KONOLIGE T , WILSON C . You are how you click:clickstream analysis for sybil detection[C]// The 22rd USENIX Security Symposium. Washington,USA, 2013: 241-256.
[15]	CAO Q , YANG X , YU J . Uncovering large groups of active malicious accounts in online social networks[C]// The 2014 ACM SIGSAC Conference on Computer and Communications Security. Scottsdale,USA, 2014: 477-488.
[16]	LESKOVEC J , FALOUTSOS C . Sampling from large graphs[C]// The 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,Philadelphia. USA, 2006: 631-636.
[17]	GJOKA M , KURANT M , BUTTS C . Walking in facebook:a case study of unbiased sampling of OSNs[C]// IEEE Proceedings of INFOCOM. California,USA, 2010: 1-9.
[18]	HU J , BAGGA A . Categorizing images in Web documents[J]. MultiMedia,IEEE, 2004,11(1): 22-30.
[19]	WAN C , GENG J , CHENG H . Image spam identifying algorithm based on color and corner feature[J]. Computer Engineering, 2009,35(15): 209-211.
[20]	DREDZE M , GEVARYAHU R , ELIAS B A . Learning fast classifiers for image spam[C]// In Conference on Email and Anti-Spam(CEAS). California,USA, 2007: 1-9.
[21]	GONEN M , ALPAYDIN E . Supervised learning of local projection kernels[J]. Neurocomputing, 2010,73(10): 1694-1703.
[22]	CORTES C , VAPNIK V . Support-vector networks[J]. Machine learning, 1995,20(3): 273-297.
[23]	KUBAT M , MATWIN S . Addressing the curse of imbalanced training sets:one-sided selection[C]// International Conference on Machine Learning. Tennessee,USA, 1997: 179-186.
[24]	e1071:misc functions of the department of statistics,probability theory group[EB/OL]. . 2015.10.11.
[25]	Rpart:recursive partitioning and regression trees[EB/OL]. . 2015.10.11.
[26]	RandomForest:breiman and cutler's random forests for classification and regression[EB/OL]. . 2015.10.11.
[27]	Nnet:feed-forward neural networks and multinomial log-linear models[EB/OL]. . 2015.10.11.

相册	训练集	测试集	总数
Spam	825	353	1 178
正常	825	353	1 178

真实值		预测值
真实值	Spam相册		正常相册
Spam相册	100.0%		0.0%
正常相册	1.8%		98.2%

相册	准确率	召回率	F₁值
Spam	0.982	1.000	0.991
正常	1.000	0.982	0.991

分类器		准确率			召回率			F₁值
分类器	Spam相册		正常相册	Spam相册		正常相册	Spam相册		正常相册
SVM	1.000		0.982	1.000		0.982	0.991		0.991
Na?ve Bayes	0.982		0.970	0.970		0.982	0.976		0.976
Decision Tree	0.988		0.988	0.988		0.988	0.988		0.988
Random Forest	0.982		0.994	0.994		0.982	0.988		0.988
Neural Network	0.960		0.988	0.988		0.960	0.973		0.973

排名	卡方检验	信息增益
1	照片描述URL比率	照片描述URL比率
2	相册中照片数	相册中照片数
3	颜色均值—红色	颜色均值—红色
4	颜色均值—绿色	颜色均值—绿色
5	颜色均值—蓝色	颜色均值—蓝色
6	相册存在时间	相册存在时间
7	色饱和度均值	色饱和度均值
8	主色覆盖范围均值	主色覆盖范围均值
9	照片描述比率	相册标题长度
10	相册标题长度	照片描述比率

在线社交网络中Spam相册检测方案

Detecting Spam albums in online social network

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 27

相关文章 1

Metrics

推荐阅读 0

数据	值
总账号数	49 156
总相册数	315 115
Spam用户	5 242
正常用户	43 914
Spam相册	89 163
正常相册	225 952