论文部分内容阅读
提出了一种基于高斯随机乘法的社交网络隐私保护方法.该算法利用无向有权图表示社交网络,通过高斯随机乘法来扰乱其边的权重,保持网络最短路径不变并使其长度应与初始网络的路径长度尽可能接近,以实现对社交网络的隐私保护.从理论上证明了算法的可行性及完美算法的不存在性.采用这种随机乘法得到的仿真结果符合理论分析结果.
This paper proposes a social network privacy protection method based on Gaussian random multiplication.The algorithm uses the undirected right graph to represent the social network and uses Gaussian random multiplication to disturb the weight of its edge and keep the shortest path of the network unchanged The path length of the initial network is as close as possible to realize the privacy protection of the social network, and the feasibility of the algorithm and the non-existence of the perfect algorithm are proved theoretically.The simulation results obtained by this randomized multiplication accord with the theoretical analysis results.