KDVEM: a k-degree anonymity with vertex and edge modification algorithm
KDVEM: a k-degree anonymity with vertex and edge modification algorithm
复制标题
KDVEM：具有顶点和边修改算法的 k 度匿名
DOI：
10.1007/s00607-015-0453-x
复制
发表时间：
2015-12-01
期刊：
影响因子：
3.7
通讯作者：
Al-Rodhaan, Mznah
中科院分区：
文献类型：
--
作者：
Ma, Tinghuai;Zhang, Yuliang;Al-Rodhaan, Mznah
隐私是社交网络数据共享中最重要的问题之一。结构匿名化是防止用户因图修改而被重新识别的有效方法。匿名化后扭曲的图结构的数据利用是一个非常严重的问题。减少效用损失是一种新的衡量标准，而k-匿名性则是保证隐私保护的一个标准。现有的使用顶点度修改的匿名化算法通常会对原始社交网络图引入大量的失真。在本文中，我们提出了一种带有顶点和边的匿名度修改算法，该算法包括两个阶段：首先，找到每个顶点的最佳目标度；其次，确定增加顶点度的候选者并添加顶点之间的边以满足要求。社交网络的社区结构因素和顶点之间的路径长度用于评估匿名化方法。对现实世界数据集的实验结果表明，匿名数据和原始数据之间的平均相对性能是我们的方法中最好的。
Privacy is one of the most important issues in social social network data sharing. Structure anonymization is a effective method to protect user from being reidentfied through graph modifications. The data utility of the distorted graph structure after the anonymization is a really severe problem. Reducing the utility loss is a new measurement while k-anonymity as a criterion to guarantee privacy protection. The existing anonymization algorithms that use vertex's degree modification usually introduce a large amount of distortion to the original social network graph. In this paper, we present a -degree anonymity with vertex and edge modification algorithm which includes two phase: first, finding the optimal target degree of each vertex; second, deciding the candidates to increase the vertex degree and adding the edges between vertices to satisfy the requirement. The community structure factors of the social network and the path length between vertices are used to evaluated the anonymization methods. Experimental results on real world datasets show that the average relative performance between anonymized data and original data is the best with our approach.