Research On Data-sharing Oriented Lineage Privacy Preserving Techniques

With thriving of data sharing,demands of the data lineage publishing become increasingly urgent.The most common form of data lineage is lineage workflow.The lineage workflow describes about how data is generated and evolves with time.Some functional modules and in-out mapping relations in the lineage workflow may contain sensitive information about the owner.It inevitably brings the privacy protection problem to sharing the lineage workflow.How to protect the privacy in the lineage workflow publishing has become one of the research hotspots.This article focuses on the key technologies for privacy-preserving independent lineage workflow publishing and privacy-preserving data and lineage workflow collaborative publishing.The thesis work is as follows:(1)?-privacy model has the disadvantages of limited attribute domains,difficulty in set ? value and unreasonable attributes selection strategy.Concerning these problems,a novel privacy-preserving lineage workflow model ?-GLprivacy is devised to balance the tradeoff between privacy-preserving and utility of the lineage workflow.The devised model applies the generalization and introduces the generalized level.Furthermore,an effective privacy-preserving lineage workflow publishing method GPPub based on generalization is proposed to achieve the privacy security in the lineage workflow publishing.(2)Concerning the scenario that business data is published prior to its lineage workflow,a novel privacy-preserving lineage workflow model(?,?)-POMS is devised.The internal correlation mechanism between the lineage workflow and data is constructed by introducing the anonymous domain and applying the decision tree algorithm.Furthermore,an effective privacy-preserving lineage publishing method MBPub based on(?,?)-POMS model is proposed to achieve the privacy security in the lineage workflow publishing while achieving the privacy security in the data publishing.The concept of module fuzziness was introduced to measure the degree of module privacy disclosure,and the restricted publishing strategy was adopted to destroy the mapping relationship of modules,so as to maintain the stability of the structure of the lineage workflow while protecting the privacy security of modules.Theoretical analysis and experimental results show the proposed methods can maintain the availability of structured queries of the lineage workflow while ensure the privacy security of lineage workflow.