Supervised classification method based on hybrid neural network

A hybrid neural network, supervised classification technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as the inability to effectively use features, improve classification accuracy, excellent classification accuracy, and solve heterogeneity. sexual effect

Inactive Publication Date: 2019-07-05
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] (1) Tree-based models (such as decision trees and random forests) usually can only be divided according to the information gain of a single attribute when dividing data, but cannot effectively use the relationship between features

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Supervised classification method based on hybrid neural network
  • Supervised classification method based on hybrid neural network
  • Supervised classification method based on hybrid neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] In order to better illustrate the purpose and advantages of the present invention, below in conjunction with the attached figure 1 The implementation manner of the method of the present invention is described in further detail with embodiment.

[0021] The specific process is:

[0022] Step 1, assuming that heterogeneous data can be divided into K subsets of homogeneous data, given n labeled samples, is the i-th sample in the j-th component, for the data set {x i ,y i} Perform K-means clustering to obtain K subsets, and initialize the parameter g according to the clustering results ij , if sample x i is assigned to the jth subset, then g ij = 1, otherwise g ij =0.

[0023] Step 2, use the samples in each subset to train K NN models, and get the parameter β of the NN model j .

[0024] Step 3, using the EM algorithm to jointly optimize the gating function and the local NN model, combined with the attached figure 2 The specific implementation method is describ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a supervised classification method, in particular to a supervised classification method based on a hybrid neural network, and belongs to the technical field of computers and information science. The method comprises the following steps of completing initialization of data division by using a K-means algorithm; training a local NN model; using an EM algorithm to jointly optimize the gating function and the expert model; updating the gate control network parameters through SGD, re-dividing the data, and using the newly-divided data subsets for retraining the local NN model; repeating the above steps until convergence. According to the supervised classification method provided by the invention, the hybrid neural network is fused to supervised classification, so that the data with heterogeneity are divided into a plurality of homogeneous data subsets, a local expert model is learned in each subset, the classification accuracy is effectively improved, and the classification accuracy is superior to that of most other supervised classification algorithms.

Description

technical field [0001] The invention relates to a supervised classification method, namely a supervised classification method based on a mixed neural network, which belongs to the technical field of computer and information science. Background technique [0002] Classification, as an important part of supervised learning problems, has attracted much attention for its various applications. There are currently many classification models such as Logistic Regression, Cox Regression, C4.5 Decision Trees, Support Vector Machines (SVM) and Neural Networks. However, these models are single models that can only fit the data globally, and most of the data are always affected by unobservable factors during the collection process, so they are heterogeneous. When the data is heterogeneous, potential unobservable factors make the data have different distributions, so a single model is inevitably affected by different data distributions. [0003] In contrast, hybrid models are usually us...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/23213G06F18/241
Inventor 罗森林王殿元潘丽敏胡雅娴
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap