Social network data differential privacy protection method based on uncertain graph

• Main Authors: WANG LI'E, LIU PENG, JIANG QUAN, XU YUANXIN, LI XIANXIAN, FU XINGCHENG
• Format: Patent
• Language: Chinese; English
• Published: 17.04.2018
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The present invention discloses a social network data differential privacy protection method based on an uncertain graph. The method comprises: constructing an original graph, and constructing an uncertain graph with the weight value and an adjacency matrix of the uncertain graph according to the original graph; according to the uncertain graph, constructing a noise adjacency matrix which is required to be added by satisfying the differential privacy; and adding the adjacency matrix with the noise adjacency matrix of the uncertain graph to obtain the adjacency matrix of a to-be-published graph, converting the adjacency matrix of the to-be-published graph into a social network graph for publishing. According to the method disclosed by the present invention, when adding the noise, not only satisfying differential privacy is ensured, and more structure information of the original graph can be saved, so that the published graph will not add excessive noise, and the data analyst can do moreresearch and analysis in