A Classification Method for Radar Chart Representation of Numerical Data Based on Convolutional Neural Networks

Abstract

The present invention provides a classification method based on the radar chart representation of numerical data based on convolutional neural network, the steps are as follows: perform feature sorting on numerical data; use the data after feature sorting to select the optimal feature combination; use the radar chart to The optimal feature combination numerical data is represented as image data; the basic structure of the convolutional neural network is constructed; the convolutional neural network model is trained. The present invention converts the numerical data into image data through the radar chart representation method, and at the same time retains the information between the data as much as possible, so as to make full use of the powerful feature extraction ability of the convolutional neural network, and convert the numerical data in the image data into image data. The topological structure features and deeper features are extracted for classification, which effectively improves the classification accuracy of numerical data compared with traditional data-driven classification methods.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, relates to a convolutional neural network classification technology, and specifically refers to a classification method based on a radar chart representation of numerical data of a convolutional neural network. Background technique [0002] As an image recognition method, convolutional neural networks have attracted extensive attention in recent years. It is a deep feed-forward neural network whose essence is to build a large number of filters to extract the features of the input data. As the network gets deeper, the extracted features become more and more abstract. Finally, the input data is represented as a sequence of abstract features that have been translated, rotated and scaled multiple times. In addition, convolutional neural networks are characterized by sparse connections, weight sharing, and spatial or temporal subsampling. Sparse connections establish spatial relationshi...

AI Technical Summary

Problems solved by technology

The traditional data-driven method is difficult to extract the characteristics of some faults, making it difficult to detect the fault, and the overall fault detection rate is low

Images