Unlock instant, AI-driven research and patent intelligence for your innovation.

Hierarchical labeled sample-oriented hidden multi-label classification method

A technology of labeling samples and classification methods, which is applied to instruments, character and pattern recognition, computer components, etc., can solve problems such as unnatural transformation results and affect the learning effect of multi-label classifiers, and achieve reasonable encoding rules and decoding rules. Effect

Pending Publication Date: 2021-09-03
SOUTHEAST UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this classification method can exploit the correlation between all labels, the existing direct conversion strategy of converting hierarchical classification into multi-label classification will cause unnatural conversion results, which in turn affects the learning effect of multi-label classifiers

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
  • Hierarchical labeled sample-oriented hidden multi-label classification method
  • Hierarchical labeled sample-oriented hidden multi-label classification method
  • Hierarchical labeled sample-oriented hidden multi-label classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. It should be noted that the words "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to the directions in the drawings, and the words "inner" and "outer ” refer to directions towards or away from the geometric center of a particular part, respectively.

[0051] Such as figure 1As shown, a hidden multi-label classification method for hierarchically labeled samples in this embodiment includes the following steps:

[0052] (1) The user selects a training sample from the storage device;

[0053] (2) According to the training sample, extract the feature set, and label the hierarchical mark set;

[0054] Step (2) According to the ...

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 discloses a hierarchical labeled sample-oriented hidden multi-label classification method. The method comprises the following steps: (1) selecting a training sample from a storage device by a user; (2) extracting a feature set according to the training sample, and marking a level mark set; (3) coding the hierarchical mark set of the training sample into a hidden multi-mark set according to a coding rule; (4) learning a hidden multi-label classifier according to the feature set of the training sample and the hidden multi-label set; (5) classifying the test samples by using a hidden multi-label classifier; (6) decoding the hidden multi-mark set of the test sample into a hierarchical mark set according to a decoding rule; (7) if the user is satisfied with the classification result, ending the process; otherwise, selecting more training samples from the storage device, and executing the step (2). The method is suitable for classifying the samples taking the mark paths in the mark tree as the mark set.

Description

technical field [0001] The invention relates to a classification method, which is mainly aimed at samples with a tree-like hierarchical marking system, belongs to the technical field of hierarchical classification, and in particular relates to a hidden multi-label classification method for hierarchically marked samples. Background technique [0002] Hierarchical classification is a task of classifying samples with a hierarchical labeling system. Assume that the complete set of markers C = {c 1 ,...,c l}, there exists a binary relation <: [0003] (1) For any mark c i ∈C, satisfying [0004] (2) For any two marks c i , c j ∈C, if c i <C j ,So [0005] (3) For any three markers c i , c j , c k ∈C, if c i <C j and c j <C k , then c i <C k . [0006] (4) There is a label r∈C, and for any label c≠r, c<r is satisfied. [0007] Then the marker corpus C constitutes a hierarchical marker, where marker r is the root. [0008] In general, a la...

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
IPC IPC(8): G06K9/62
CPCG06F18/2431G06F18/2415
Inventor 张敏灵於泽邦
Owner SOUTHEAST UNIV