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

A multi-label classification processing method and device for incompletely labeled data

A technology for labeling data and processing methods, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problem of low practicability of label classification, and achieve the effect of easy implementation and high practicability

Active Publication Date: 2021-08-03
北京京航计算通讯研究所
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of the above analysis, the embodiment of the present invention aims to provide a multi-label classification processing method and device for incompletely labeled data, to solve the problem that the existing multi-label classification method is less practical for label classification with overlapping data. question

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
  • A multi-label classification processing method and device for incompletely labeled data
  • A multi-label classification processing method and device for incompletely labeled data
  • A multi-label classification processing method and device for incompletely labeled data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] Preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of the application and together with the embodiments of the present invention are used to explain the principle of the present invention and are not intended to limit the scope of the present invention.

[0050]In the field of network security, the traffic data received by different IP addresses represents the behavior distribution of users. Due to the uniqueness of user services, users of different services will generate different traffic distributions. These users will generate some common operations in the terminal. In the process of collecting user data, collecting the data of a specific IP address will generate an automatic label, that is, the user of the IP address. If the user is used as a category, it is a classification problem to identify the user through the traffic data generated by...

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 multi-label classification processing method and device for incompletely labeled data, which belongs to the technical field of data classification and solves the problem that the existing multi-label classification method has low practicability for classifying labels with overlapping data. The method includes: obtaining a training data set; performing parameter training on the neural network based on the training data set to obtain a trained neural network; preprocessing the output of the trained neural network intermediate layer to obtain an optimal network structure of the neural network; The incompletely labeled data to be classified is input into the optimal network structure of the neural network, and all label category vectors corresponding to the incompletely labeled data to be classified are obtained. The multi-label classification of incompletely labeled data is realized, the accuracy of data label classification is improved, and it has high practicability.

Description

technical field [0001] The invention relates to the technical field of data classification, in particular to a multi-label classification processing method and device for incompletely labeled data. Background technique [0002] In the traditional data classification problem, the data of each category has more significant category characteristics, and there is no overlap between different categories, that is, the same data cannot have the characteristics of two categories at the same time. Therefore, there is a classification hypersurface in theory, so that the data of different categories are distributed in different regions separated by the classification hypersurface. [0003] In some application fields, due to specific scenarios and data identification methods, different types of data have a certain degree of overlap, and the distribution of different types of data has both common parts and unique data. For example, in the field of network security, traffic data received...

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 Patents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/23G06F18/22G06F18/2431
Inventor 王础王佳佳陈浩侯健王凤敏王立霞李娜
Owner 北京京航计算通讯研究所