摘要
随着社交平台的发展,社交媒体网络逐渐成为攻击者进行僵尸网络渗透的理想平台。社交僵尸网络利用社交平台自动化程度高、灵活性强与普及度高等特性构建隐蔽信道进行通信,以达到窃取社交平台用户信息、散布不良信息污染网络环境、引导控制舆论等目的。传统的僵尸网络检测机制无法有效地检测社交僵尸网络,为社交媒体的安全性带来极大的挑战。从社交僵尸网络的概念入手,阐述社交僵尸网络在不同社交平台上的发展脉络和发展趋势,研究不同社交媒体上的社交僵尸网络攻击原理和群体特征以及隐蔽型社交僵尸网络的隐蔽手段。在此基础上,将社交僵尸网络的检测方法分为服务器端检测方法和客户端检测方法,并对近年来出现的基于隐写技术和基于机器学习的检测方法进行分析,同时给出社交僵尸网络的反制技术和接管方法的研究现状及发展思路,并对该领域的未来研究方向进行展望。
With the development of social platforms,social media networks have gradually become an ideal platforms for attackers to infiltrate Botnets.Social botnets utilize the high flexibility,degree of automation,and popularity of social platforms to create covert channels for communication to steal social media platform user information,disseminate bad information to pollute the network environment,and control public opinion.Traditional botnet detection mechanisms cannot effectively detect social botnets,causing significant challenges to social media security.From the concept of social botnets,the developmental context and trends of social botnets on different social platforms are explained.In addition,the attack principles and group characteristics of social botnets on different social media and the concealment methods of covert social botnets are analyzed.On this basis,the detection methods of social botnets are divided into server-side and client-side,and the detection methods based on steganography and machine learning in recent years are analyzed.Furthermore,the research status and development ideas of the countermeasure technology and takeover method of social botnets are discussed,and future research directions in this field are suggested.
作者
葛昕
邹福泰
郭万达
谭越
李林森
GE Xin;ZOU Futai;GUO Wanda;TAN Yue;LI Linsen(Information Office,University of Shanghai for Science and Technology,Shanghai 200093,China;School of Electronic Information and Electrical Engineering,Shanghai JiaoTong University,Shanghai 200240,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2022年第8期12-24,共13页
Computer Engineering
基金
国家重点研发计划(2018YFB0803503)。
关键词
网络安全
社交僵尸网络
命令与控制信道
隐写技术
机器学习
cyber security
social Botnets
Command and Control(C&C)channels
steganography
machine learning