Privacy protection multi-mechanism data classification method based on homomorphic encryption

Abstract

The invention discloses a privacy protection multi-mechanism data classification method based on homomorphic encryption. Firstly, a user generates a screening condition of training data and sends thescreening condition to a computing center; a computing center generates encryption parameters by using a homomorphic encryption algorithm and sends the encryption parameters to each data providing mechanism; each data providing mechanism screens the local data marked with the classification tags to obtain training data, generates respective public key and private key by utilizing a homomorphic encryption algorithm according to the encryption parameters, encrypts the training data through the public key and sends the training data to the computing center; a computing center and each data providing mechanism jointly execute logistic regression analysis under homomorphic encryption to obtain a data classification model and send the data classification model to a user; and a user inputs the to-be-classified data into the data classification model to obtain a classification result. Compared with the prior art, the method has higher safety in practical application, personal privacy is effectively protected, benefit loss possibly generated by a data providing mechanism is avoided, and development of multi-mechanism data analysis is promoted to a great extent.

Description

technical field [0001] The invention belongs to the technical field of multi-institutional data analysis, and in particular relates to a privacy-protecting multi-institutional data classification method based on homomorphic encryption. Background technique [0002] Currently, most data analysis research is conducted on limited datasets from a single institution. However, such an approach has significant limitations. On the one hand, for some specific research, a single institution may not be able to generate enough supporting data, for example, for a rare disease, a single medical institution usually cannot provide enough data; on the other hand, due to the limitations of various observation techniques With the development, more and more heterogeneous data have been generated, which also poses more challenges to data mining. For example, the research by Joshua C Denny, a scientist at Vanderbilt University in 2013, showed that in a single medical institution's electronic Ge...

AI Technical Summary

Problems solved by technology

On the one hand, through proper processing of these exposed information, some sensitive information can be deduced, leading to the disclosure of personal privacy

On the other hand, data itself has value, and a large amount of exposed information may be maliciously used by third parties without authorization, resulting in the loss of interests of data providers

This significantly hampers multi-institutional data analysis

Images