Information cyber-physic system data classification method on basis of genetic program decision-making trees

Abstract

The invention relates to an information cyber-physic system data classification method on the basis of genetic program decision-making trees, which comprises the following steps: firstly, carrying out data collection by using a physical unit; then transmitting the collected data to an information unit; finally, carrying out further analysis and processing in the information unit. The operation of the information unit comprises the following steps that 1, the information unit adopts a mode with replacement to establish a plurality of decision-making trees and records a category number of the data; 2, the established decision-making trees are tested and classification accuracy and cost loss of each decision-making tree; 3, a genetic program is adopted to carry out optimization processing on all the decision-making trees in the information unit and finally, a number of decision-making trees, which is equal to the category number of the data, are kept back; 4, the left decision-making trees are utilized to modify training data categories on the basis of reducing cost loss and a final decision-making tree is generated by the modified training data. When a certain classification accuracy and low cost loss are ensured, time consumption of constructing the final decision-making tree is effectively reduced.

Description

technical field [0001] The invention relates to a decision tree classification method based on genetic programming, which is mainly used to solve the problems of cumbersome classification process, excessive time consumption and low efficiency when considering the cost loss of large-scale data in a cyber-physical fusion system, and belongs to the cyber-physical fusion system The cross-technical application field of data mining and data mining. Background technique [0002] Cyber-physical fusion system is considered to be the third wave of information technology in the world after computers and the Internet. The cyber-physical fusion system can be understood as an efficient networked intelligent information system based on embedded devices. It is a next-generation intelligent system with highly independent perception, independent judgment, self-regulation and self-government capabilities, and can realize the interconnection and collaboration between the virtual world and the r...

AI Technical Summary

Problems solved by technology

[0010] Technical problem: The purpose of this invention is to provide a data classification method for cyber-physical fusion system based on genetic programming decision tree. When the cost loss of data is considered, the classification process is cumbersome, time-consuming, and inefficient

Images