摘要
随着互联网技术的发展和智能终端的普及,社交网络中产生了大量用户隐私数据,公开发布社交网络数据将提高用户隐私泄露的风险,需要对数据进行匿名化处理然后进行发布。传统社交网络k度匿名方法在图数据连续发布中的匿名方式,存在大量冗余计算及无法抵抗度时序推理攻击的问题,为此,提出一种连续发布图数据的改进k度匿名算法。通过定义度时序矩阵来一次性地构建满足k匿名性要求的k度时序矩阵,在k度时序矩阵的基础上提取不同时刻的k度向量,将其作为时刻图的匿名向量,通过图修改方法对前一时刻的匿名图进行处理,得到后续一系列的匿名图版本,从而缩短每一次重新匿名所消耗的时间,同时抵抗基于度变化实现的度时序背景知识攻击。在真实社交网络数据集上进行实验,结果表明,相对kDA算法,该算法的总体运行效率以及网络结构属性可用性均较优。
With the development of Internet technology and popularity of intelligent terminals,a large amount of user privacy data have been generated in social networks.The public release of social network data increases the risk of user privacy disclosure.Therefore,anonymizing the data is necessary before publishing it.The traditional k-degree anonymity method in the continuous publishing of graph data requires numerous redundant calculations and cannot resist temporal reasoning attack.Therefore,an improved k-degree anonymity algorithm for continuous publishing of graph data is proposed.A k-degree time series matrix satisfying the requirements of k anonymity is constructed by defining the degree time series matrix.From this matrix,a k-degree vectors is extracted as the anonymous vector of the time chart.The anonymous graph at the previous time is processed using the graph modification method to obtain a series of subsequent anonymous graph versions to shorten the time consumed by each anonymity.Simultaneously,it resists the degree time series background knowledge attack using the degree change.The experiments on real social network datasets show that the overall operation efficiency and network attribute availability of this algorithm are better than the kDA algorithm.
作者
朱黎明
丁晓波
龚国强
ZHU Liming;DING Xiaobo;GONG Guoqiang(College of Computer and Information Technology,China Three Gorges University,Yichang,Hubei 443002,China;Hubei Province Engineering Technology Research Center for Construction Quality Testing Equipment,Yichang,Hubei 443000,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2022年第5期154-161,共8页
Computer Engineering
基金
国家重点研发计划“网络空间安全”专项基金(2016YFB0800403)。
关键词
图数据
隐私保护
k度匿名
社交网络
时序矩阵
graph data
privacy protection
k-degree anonymity
social network
time sequence matrix