Research And Implementation Of Graph Data Anonymization Based On Deep Learning Method

With the rapid development of research and application of graph data,private information such as individual existence and associations in graph data also faces huge risks of leakage and abuse.In particular,when we publish or share data with others,we need to anonymize the data by anonymization methods.Existing graph data anonymization methods mostly specify a feature in the graph for clustering,noise addition,generalization,etc.,which is difficult to preserve privacy.And the anonymity is usually obtained by giving up data utility.Aiming at the above problems,we designed a graph anonymization method which is based on a feature learning model of Generative Adversarial Network.We used the differential privacy to ensure the privacy and take both anonymity and utility into consideration.The main work consists of the following two parts:Firstly,we designed a graph feature learning method based on Generative Adversarial Network(GAN).The method used the bias random walk strategy to sample the node sequence from graph data,and trained the GAN model.After training,the GAN generated a set of simulation sequences that are highly similar to the real sampled sequence.Finally,we demonstrated that the model had good feature learning ability through embedding visualization and link prediction experiments,compared with other anonymous graphs.Secondly,we proposed an anonymous graph construction method based on the simulation node sequence.We calculated the number of edges in the node sequences and constructed a probability adjacency matrix.The differential privacy noise is added to get the anonymous probability adjacency matrix.Finally,we extract the edges from the anonymous matrix and then constructed the anonymous graph.Through experiments such as metric evaluation,community detection,and de-anonymization attack,we proved that the anonymous method we proposed is better than the current mainstream anonymous method.