Privacy protection reissuing method for multiple sensitive attribute data

Abstract

The invention belongs to the field of information privacy protection issuing safety, and relates to a privacy protection reissuing method for multiple sensitive attribute data. The privacy protection reissuing method includes the steps of dividing reissued data into retention data and refresh data and processing the retained data and the refreshed data and respectively processing the retained data and the refreshed data on the basis of defining privacy protection diversity parameters, sensitive attribute safety change sets and record safety change sets, issuing the retained data through an L-Inclusive model, and issuing the rest of refresh data through an L-Coverage model. According to the privacy protection reissuing method, relevance among the multiple sensitive attributes, relevance among different values of each sensitive attribute and relevance among multiple times of issuing of data are taken into full consideration, privacy attack can be effectively restrained, and the function of privacy protection issuing is achieved.

Description

technical field [0001] The invention belongs to the field of information privacy protection release security, and in particular relates to a privacy protection re-release method for multi-sensitive attribute data. Background technique [0002] In recent years, the types and quantities of various information and data collections are increasing rapidly. These data contain a large amount of private information. If the information is released directly, it will cause a serious leakage of privacy. In order to ensure the security of the information, it is necessary to perform privacy protection processing while publishing the data. Published data table attributes are generally divided into three categories: explicit identifier (Identifier), which can clearly identify the identity attribute of the record; quasi-identifier QI (Quasi Identifier), which exists in both privacy tables and appearances, and can be identified by links Individual information; sensitive attribute SA (Sensiti...

AI Technical Summary

Problems solved by technology

[0004] L-diversity technology is a commonly used method of publishing privacy protection. Based on this, the multi-dimensional bucket technology is used to deal with the static publishing of multi-sensitive attributes. The m-Invariance model is a method used to deal with data republishing. However, these methods do not consider the release of dynamic multi-sensitive attributes, let alone the correlation between sensitive values ​​and the correlation between sensitive attributes.

Images