Anonymous method for missing data and storage medium

Abstract

The invention provides a missing data anonymity method and a storage medium. The missing data anonymity method comprises the steps of setting a clustering parameter k; setting l-diversity model parameter l; clustering all the data records in the data set according to the similarity judgment values of the data records in the data set, the l-diversity model parameters and the clustering parameters,dividing the data set into a plurality of clustering cluster groups, and obtaining a data set with a plurality of clustering clusters; and performing generalization processing on each clustering cluster group in the data set, performing generalization on each clustering cluster to obtain an equivalence class, and in the obtained equivalence class, the values of the data records in the same equivalence class on the quasi-identifier attribute are the same, thereby finishing anonymous processing. According to the method, the availability of the data after anonymization processing is guaranteed tothe maximum extent by processing the incomplete data set, information loss caused by a traditional anonymization method is reduced. Meanwhile, sensitive attributes related to a user are protected byl-diversity, and the safety of anonymization processing of the data set is improved.

Description

technical field [0001] The invention relates to an anonymous method for missing data, in particular to an anonymous method when incomplete data is contained in a data set. Background technique [0002] With the dissemination and sharing of a large amount of information, while bringing convenience to social development and people's lives, it also increases the risk of privacy information leakage. Although the display identifier (such as name, ID number, credit card number, etc.) will be eliminated when the data is released, the attacker can carry out link attacks through multiple data sets, which will still cause a certain degree of privacy leakage. The k-anonymity in data publishing anonymity technology and its series of optimization schemes have been extensively studied and applied to the anonymity protection of data publishing. These methods reduce the information loss caused by generalization, but at the same time increase the risk of privacy leakage. When the existing ...

AI Technical Summary

Problems solved by technology

Traditional anonymity technology removes incomplete records contained in medical data sets during data processing, resulting in excessive loss of information and affecting the normal use of data sets

Images