Data privacy protection method in hybrid cloud environment

Abstract

The invention belongs to the technical field of data privacy protection, and particularly relates to a data privacy protection method in a hybrid cloud environment, which comprises the following steps: step 1, taking a DaaS bearing platform as a hybrid cloud, and analyzing the potential risk of data privacy leakage after the introduction of a cloud platform according to the characteristics of high-dimensional sparse data and a data release mode; step 2, on the basis of an anonymous segmentation strategy, analyzing a constraint scene with maximum data availability through a greedy strategy; step 3, analyzing the influence of noise addition on data availability by utilizing statistical search of interactive differential privacy protection; and step 4, for data error segmentation caused by parallel anonymous segmentation, reducing the data volume reserved on the private cloud by sharing the aggregation cluster. According to the method, information loss can be reduced, and the execution efficiency of the algorithm is improved, so that the feasibility and practicability of privacy protection of data are improved.

Description

technical field [0001] The invention belongs to the technical field of data privacy protection, and in particular relates to a data privacy protection method in a hybrid cloud environment. Background technique [0002] Today, many milestone research results have emerged in privacy-preserving data publishing technologies. Most of these solutions are designed for structured data, and the data structures of big data are diverse, including not only traditional structured data, but also semi-structured data. Data (such as transaction data, user web browsing records, etc.), unstructured data (such as pictures, audio, video files, etc.), whether the existing privacy protection scheme for structured data can be directly migrated to the privacy protection scenario for big data , has become a hot topic of academic discussion. [0003] This type of data has the characteristics of high dimensionality and data sparseness, and privacy protection for it cannot directly copy the privacy pr...

AI Technical Summary

Problems solved by technology

First of all, due to the difference in user purchase needs, it is difficult to make a binary division of the product itself (ordinary product, sensitive product) from the perspective of the product itself; secondly, due to the explosive growth of data attribute dimensions, the reasoning and mapping relationship between attributes Intricate; third, due to the sparse distribution of the mapping of records on each attribute, the number of record equivalence classes increases sharply. If k-anonymity is to be achieved, the amount of added noise will also increase, which will seriously deteriorate the availability of data.

Images