Privacy Protection Workflow Publishing Under Differential Privacy

The workflow has been widely used in data quality assessment, error data location and other fields. As data sharing deepens, so does the need to share data lineages. The topology of the lineage workflow contains private information that includes the data generation process, that is, the privacy of the lineage workflow structure, which directly exposes the structure privacy leakage of the lineage workflow. There are the following deficiencies in the privacy protection methods of the lineage workflow structure: (1) The privacy protection method based on the restricted release has a weak theoretical foundation and can only qualitatively measure the privacy protection effect of the lineage workflow structure; (2) Focusing on the maintenance of the local mapping relationship of modules, the maintenance of the key path of the lineage workflow is weak. Aiming at the above problems, this paper proposes a privacy protection method PPWP-DP for the lineage workflow structure, which satisfies the differential privacy. Key path and key path priority concepts are introduced. On this basis, the θ-project projection algorithm is proposed to reduce the degree of the lineage workflow. At the same time, according to the user’s preference for key path priority, the maintenance of high priority key path reachability is achieved. The concept of oi-sequence is introduced to extract the structure characteristics of the lineage workflow, and add Laplacian noise to the oi-sequence to satisfy the differential privacy constraint. Adjust the global sensitivity of the oi-sequence after noise addition by the θ-project algorithm to reduce the Laplacian noise scale. Finally, the perturbed oi-sequence is used to reconstruct the lineage workflow for publication, which realizes the workflow privacy security and the maintenance of key path accessibility. Theoretical analysis and experiments verify the effectiveness of the proposed algorithm.