Privacy protection type data classification method for information physical fusion system

Abstract

The invention provides a privacy protection type data classification method for an information physical fusion system. The privacy leakage problem of the information physical fusion system in the distributed data mining process is solved through a random decision-making tree and a threshold addition homomorphic password. The method comprises the steps that firstly, the structure of the random decision-making tree is determined by all units of the information physical fusion system; secondly, node values of the decision-making tree are calculated; finally, an established random decision-making forest and a threshold addition homomorphic public key password are used for classifying new living examples. According to the privacy protection type data classification method, a classifier is accurately established through the distributed random decision-making tree, the threshold addition homomorphic password is combined, and data in the information physical fusion system can be classified under the premise that efficient and reliable privacy protection is provided.

Description

technical field [0001] The present invention provides a privacy-protected data classification method for a cyber-physical fusion system based on a random decision tree, which mainly uses a distributed random decision tree and a threshold additive homomorphic cipher to provide efficient and reliable privacy during the classification process of the cyber-physical fusion system. Protection, which belongs to the cross-technical application field of information security and data mining. Background technique [0002] The cyber-physical fusion system is a multi-dimensional complex system integrating computing, network and physical environment. Through the organic integration and in-depth cooperation of 3C (Computation, Communication, Control) technology, it realizes real-time perception, dynamic control and information service of large-scale engineering systems. The cyber-physical fusion system is an intelligent system with autonomous behavior. The cyber-physical fusion system can ...

AI Technical Summary

Problems solved by technology

[0007] Technical problem: The purpose of the present invention is to provide a privacy-protected data classification method for cyber-physical fusion systems, which combines distributed random decision trees and threshold additive homomorphic cryptography in the classification process to solve privacy problems in data mining. conservation issues

Images