Method of implementing negative survey and reconstructing positive data

Abstract

The present invention relates to a method of implementing the negative survey and reconstructing the positive data. The method comprises the steps of implementing the negative survey according to the form of the negative survey, at the same time, extracting a small number of samples to implement the positive survey corresponding to the title of the negative survey; gathering the data to obtain the negative selection matrixes of each option and a total sample number; carrying out the correlation analysis on the small number of samples after the positive survey and the negative survey to obtain a reconstruction matrix; according to a matrix constraint condition of the negative survey, adjusting the reconstruction matrix; applying the maximum likelihood estimation and the constraint condition of the negative survey to obtain an optimal reconstruction result. Based on the background knowledge and the event maximum likelihood estimation, the present invention provides a negative survey implementation and reconstruction method which can provide the reference for the real application mode of the negative survey, at the same time, and enables the negative data reconstruction accuracy to be improved.

Description

technical field [0001] The invention belongs to the field of privacy protection and security, and in particular relates to a method for implementing negative investigation based on background knowledge and reconstructing positive data based on maximum likelihood estimation. Background technique [0002] With the rapid development of computer networks and big data, people are exposed to a variety of massive data and information in daily life. Although this has brought convenience to people in some aspects, it has also caused various privacy data to a certain extent. leak. Therefore, protecting people's privacy is becoming more and more important. Negative representation of information is an effective privacy-preserving data representation, and negative investigation originates from negative representation of information. [0003] In a traditional questionnaire survey (positive survey), the participant is asked to select the option he belongs to (positive option), so the col...

AI Technical Summary

Problems solved by technology

[0004] The research on negative survey is still in its infancy, and the key problem is the reconstruction from negative data to positive data. Although the existing methods (NStoPS, NStoPS-I, NStoPS-II, etc.) can obtain results, they are only applicable to Uniform distribution, a theoretical reconstruction method, has not been tested on manually selected real data, so it may not be suitable for real negative investigations

[0005] Although negative surveys have experienced more than ten years of development since they were put forward, and more and more scholars have studied negative surveys, the research on negative surveys has always remained at the stage of theoretical and application verification.

Most negative investigations are just simulated negative investigations by computer, and a small part of the work, although real negative investigations are carried out, is mainly carried out by designing programs for negative selection, not by manual selection

[0006] At present, the selected model of the option proposed by the researchers obeys the ideal state such as uniform distribution and Gaussian distribution, but this may not be the case in the real negative survey. Therefore, the probability of each option being selected in the negative survey needs to be adjusted. Therefore, the existing The reconstruction algorithm from negative surveys to positive surveys may not be suitable for real negative surveys

Images