基于差分隐私保护的社交网络发布图生成模型

发布时间：2018-05-19 11:49

本文选题：差分隐私保护 + 社交网络　；参考：《同济大学学报(自然科学版)》2017年08期

【摘要】：社交网络在帮助人们建立社会性网络应用服务的同时,收集了大量的用户资料和敏感数据,通过分析这些数据可能泄露潜在的隐私信息.目前差分隐私保护模型对隐私泄露风险给出了严谨、定量化的表示和证明,极大地保证了数据的可用性.设计了一个满足差分隐私保护的社交网络发布图生成模型,首先通过图模型表示社交网络结构,并将原图按照节点特征分类为多个子图;然后利用四叉树方法对子图的密集区域进行划分,在树的叶子节点添加满足差分隐私保护的噪声;通过子图重构的方式,生成待发布图.最后,利用度分布、最短路径、聚类系数等统计分析方法,实验验证了该模型的可行性和有用性.
[Abstract]:Social network not only helps people to build social network application service, but also collects a large amount of user information and sensitive data, which may reveal potential privacy information by analyzing these data. At present, the differential privacy protection model gives a rigorous, quantitative representation and proof to the risk of privacy disclosure, which greatly ensures the availability of data. In this paper, a social network publishing graph generation model satisfying differential privacy protection is designed. Firstly, the social network structure is represented by the graph model, and the original graph is classified into several sub-graphs according to the node characteristics. Then using the quadtree method to divide the dense area of the subgraph, add the noise satisfying the difference privacy protection in the leaf node of the tree, and generate the image to be published by the way of reconstruction of the subgraph. Finally, the feasibility and usefulness of the model are verified by using statistical analysis methods such as degree distribution, shortest path, clustering coefficient and so on.
【作者单位】：同济大学电子与信息工程学院;
【基金】：国家“八六三”高技术研究发展计划(2015IM030300) 上海市科技创新计划(15DZ1101202) 上海市科委项目(14JC1405800) 同济大学中央高校基本科研业务费
【分类号】：TP309;TP393.09

【相似文献】