基于三元闭包的不确定社交网络隐私保护算法

Privacy Preservation Algorithms of Uncertain Graphs for Social Networks Based on Triadic Closure

当前,社交网络的使用量持续上升,在发布社交网络数据的同时对社交网络隐私的保护是目前研究的重点。目前社交网络隐私保护方法,大部分着重于对图隐私的保护,忽略了图数据效用的问题,导致图数据效用破坏严重,适用范围有限。针对以上问题,设计了一种基于三元闭包的潜在边算法。该算法首先通过三元闭包方法将社交网络演化的潜在边加入到原始图中,由此生成具有一定动态适应性的新图;然后使用(k,ε)-模糊算法向图中注入不确定性,生成不确定图,利用不确定图的特点达到了在保证一定隐私水平的情况下能够保持较高的图数据效用。最后,通过实验对比证明该方法隐私保护效果较好,算法适用范围较广。

The use of social networks continues to rise at present,the current research focuses on the protection of social net-work privacy while releasing social network data.At present,most of the privacy protection methods of social networks focus on the protection of graph privacy.The problem of graph utility is ignored,which leads to the serious damage of graph utility and limited ap-plication scope.To solve these problems,a potential edge algorithm based on ternary closure is proposed.In this algorithm,the po-tential edges of social network evolution are first added to the original graph through ternary closure method,and the resulting new graph will have certain dynamic applicability.Then the(k,ε)-obfuscation algorithm is used to inject uncertainty into the graph to generate uncertain graphs.By using the characteristics of uncertain graph,the utility of graph data can be maintained at a certain level of privacy.Finally,the experimental comparison proves that the method has better privacy protection effect and is applicable to a wide range.