Research On Provenance Workflow Sharing Technology For Structural Privacy Protection

The data provenance is used to describe data generation,evolution processes,and data source information,which plays an increasingly important role in the field of data quality assessment,data traceability,and information security.With people's increasing requirements for data quality and traceability,the demand for provenance sharing is more urgent,and provenance workflow,as the main form of data provenance,inevitably brings privacy leakage issues during the sharing release process.Aiming at the shortcomings of the existing provenance workflow privacy protection technology,this thesis proposes a high-strength privacy protection method that can effectively maintain workflow timing constraints and topology availability.The main work of the thesis is as follows:(1)In view of the problem that the differential privacy based d K-graph anonymous privacy technology,which maintains the weak usability of provenance workflow node flow characteristics and overall workflow timing dependency,a differential privacy based Workflow Privacy-Preserving DP?WPP is proposed.The Input/output degree d K model is introduced to describe the characteristics of workflow node degree and direction;using differential privacy mechanism to achieve differential protection of workflow summary features;meanwhile,in the process of workflow reconstruction,the Before and After sequence structure is introduced,realizing the reconstruction workflow to meet the differential privacy constraints while taking into account the effective maintenance of timing characteristics.(2)In view of the problem that existing graph structure privacy protection methods have a weaker maintenance effect on the topology structure of the workflow that characterizes the timing characteristics,on the basis of the k-degree anonymous model,an anonymous publishing method Tm WAP is proposed to maintain the workflow topology availability.By introducing a hierarchical model based on topological sequence,the effective extraction of workflow topological timing characteristics is achieved,and a topology loss function is designed to measure the availability of topological timing characteristics;by constructing a workflow minimal topological sub graph based on a hierarchical model and designing a heuristic strategy based on the topology loss function,the optimal graph operation is selected during the k-degree anonymization process,which achieves the effect of taking into account the k-degree anonymous privacy protection of the workflow and the maintenance of the usability of the topology.Theoretical analysis and experimental results show that the proposed method achieves measurable workflow privacy protection while taking into account the efficient maintenance of workflow availability.