摘要
为了有效地防范社交网络环境中用户数据隐私泄露的风险,提出一种基于聚类的社交网络保护机制。首先将全同态加密技术引入社交共享数据中,保护社交过程中数据的隐私安全;其次,在保证用户信息和数据安全的前提下构建社交网络结构图,将经过同态加密处理的用户信息和数据按照节点相似度进行聚类,并对聚类后的社交网络结构进行分类和区分;最后对聚类后的超节点进行匿名化处理和分析,结果表明该社交网络模型可以降低信息损失度,同时保证了用户数据的隐私安全,验证了数据可用性。
In order to effectively prevent the risk of user data privacy leakage in social network environment,a social network protection mechanism based on clustering is proposed.First,we introduce fully homomorphic encryption technology into social shared data to protect the privacy security of data in the social process;secondly,construct a social network structure diagram under the premise of ensuring the security of user information and data,cluster the user information and data processed by homomorphic encryption according to the node similarity,and classify and distinguish the clustered social network structure;finally,the clustered supernodes are anonymized and analyzed.The results show that the social network model can reduce the degree of information loss,ensure the privacy security of user data,and verify the availability of data.
作者
李秋贤
周全兴
LI Qiuxian;ZHOU Quanxing(Kaili University,Kaili 556011,China)
出处
《现代信息科技》
2022年第16期168-170,共3页
Modern Information Technology
基金
黔东南州科技计划项目(黔东南科合J字[2021]39号)
扶持市(州)高校质量提升工程项目(院办发[2022]10号-32)
贵州省普通高等学校青年科技人才成长项目(黔教合KY字[2020]179,黔教合KY字[2020]180)
凯里学院专项课题(XTYB1602)。
关键词
社交网络
聚类
全同态加密
匿名化
信息损失度
social network
clustering
fully homomorphic encryption
anonymization
degree of information loss
