基于差分隐私的权重社会网络隐私保护  

Privacy preserving based on differential privacy for weighted social networks

作　　者：兰丽辉[1,2] 鞠时光[1] 

机构地区：[1]江苏大学计算机科学与通信工程学院,江苏镇江212013 [2]沈阳大学信息工程学院,辽宁沈阳110044

出　　处：《通信学报》2015年第9期145-159,共15页Journal on Communications

基　　金：国家自然科学基金资助项目(61003288;61111130184);国家教育部博士点基金资助项目(20093227110005);江苏省普通高校研究生科研创新计划基金资助项目(CX10B_006X)~~

摘　　要：针对权重社会网络发布隐私保护中的弱保护问题,提出一种基于差分隐私模型的随机扰动方法可实现边及边权重的强保护。设计了满足差分隐私的查询模型-WSQuery,WSQuery模型可捕获权重社会网络的结构,以有序三元组序列作为查询结果集;依据WSQuery模型设计了满足差分隐私的算法-WSPA,WSPA算法将查询结果集映射为一个实数向量,通过在向量中注入Laplace噪音实现隐私保护;针对WSPA算法误差较高的问题提出了改进算法-LWSPA,LWSPA算法对查询结果集中的三元组序列进行分割,对每个子序列构建满足差分隐私的算法,降低了误差,提高了数据效用。实验结果表明,提出的隐私保护方法在实现隐私信息的强保护同时使发布的权重社会网络仍具有可接受的数据效用。Focusing on the weak protection problems in privacy preservation of weighted social networks publication, a privacy preserving method based on differential privacy was put forward for strong protection of edges and edge weights. The WSQuery query model was proposed meeting with differential privacy on weighted social networks, could capture the structure of weighted social networks and returned the triple sequences as the query result set. The WSPA algorithm was designed according to the WSQuery model, could map the query result set into a real mmaber vector and injected Laplace noise into the vector to realize privacy protection. The LWSPA algorithm was put forward because of the high error of the WSPA algorithm, partitioned the triples sequence of the query results into multiple subsequences, constructed the algorithms for each subsequence according with differential privacy and reduced the error and improved the data util- ity. The experimental results demonstrate that the proposed method can provide strong protection for privacy information, simultaneously the utility of the released weighted social networks is still acceptable.

关 键 词：权重社会网络 隐私保护 差分隐私 查询模型 LAPLACE分布 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]