Enterprise text multi-label labeling method and system based on attention mechanism

A multi-label and attention-based technology, applied in the field of data processing, can solve the problems of inaccurate industry classification, high labeling cost, and weak generalization ability, etc., to achieve comprehensive enterprise text data, efficient labeling process, and reduce subjective differences sexual effect

Pending Publication Date: 2022-07-08
SHANDONG COMP SCI CENTNAT SUPERCOMP CENT IN JINAN +2
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] At present, most of the industry labeling and classification schemes of enterprises rely on manual data labeling and classification, so personal subjective factors will be mixed into the classification. With the increasing number of enterprises, there will still be problems such as high labeling costs and incomplete coverage.
At this stage, some traditional machine learning methods are used, but the effect of traditional classification methods is not ideal, and the generalization ability is not strong
There are also a small number of people who use deep learning classification methods, but only for the business scope of the company, resulting in the inability to fully reflect the true state of the company and the inaccurate classification of the company's industry

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
  • Enterprise text multi-label labeling method and system based on attention mechanism
  • Enterprise text multi-label labeling method and system based on attention mechanism
  • Enterprise text multi-label labeling method and system based on attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] This embodiment provides a multi-label labeling method for enterprise text based on an attention mechanism;

[0035] like figure 1 As shown, the attention mechanism-based multi-label annotation method for enterprise text includes:

[0036] S101: Obtain basic attribute information of a data object demander;

[0037] S102: Preprocess the acquired information;

[0038] S103: Label the preprocessed data by using the trained enterprise text multi-label labeling model to obtain multiple labeling labels;

[0039] Among them, the enterprise text multi-label annotation model adopts the attention mechanism layer to extract text syntax and semantic features.

[0040] Among them, the data object demander refers to the enterprise; the data object refers to multiple labels of the enterprise text. Multiple labels, such as: mining, manufacturing, electricity, construction, wholesale and retail, transportation, accommodation and catering, information transmission, finance, real esta...

Embodiment 2

[0092] This embodiment provides an attention mechanism-based enterprise text multi-label labeling system;

[0093] A multi-label annotation system for enterprise text based on attention mechanism, including:

[0094] an acquisition module, which is configured to: acquire basic attribute information of the data object demander;

[0095] a preprocessing module, which is configured to: preprocess the acquired information;

[0096] a labeling module, which is configured to: label the preprocessed data by using the trained enterprise text multi-label labeling model to obtain a plurality of labeling labels;

[0097] Among them, the enterprise text multi-label annotation model adopts the attention mechanism layer to extract text syntax and semantic features.

[0098] It should be noted here that the above acquisition module, preprocessing module, and labeling module correspond to steps S101 to S103 in Embodiment 1, and the above modules and corresponding steps implement the same ex...

Embodiment 3

[0102]This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are Stored in the memory, when the electronic device runs, the processor executes one or more computer programs stored in the memory, so that the electronic device executes the method described in the first embodiment.

[0103] It should be understood that, in this embodiment, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors, DSPs, application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices. , discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or 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 an attention mechanism-based enterprise text multi-label labeling method and system. The method comprises the following steps: acquiring basic attribute information of a data object demander; the obtained information is preprocessed; labeling the preprocessed data by adopting a trained enterprise text multi-label labeling model to obtain a plurality of labeled labels; according to the enterprise text multi-label labeling model, text syntactic and semantic feature extraction is carried out by adopting an attention mechanism layer. According to the enterprise text label labeling method and system, enterprise texts and labels can be automatically labeled, classified and stored, a user can conveniently and accurately inquire enterprise operation content, related personnel can conveniently master the industry distribution dynamic state of local enterprises in real time, the whole process does not need manual intervention, and the system automatically completes the whole process.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a method and system for enterprise text multi-label labeling based on an attention mechanism. Background technique [0002] The statements in this section merely provide background related to the present disclosure and do not necessarily constitute prior art. [0003] At present, most of the industry labeling and classification schemes of enterprises rely on manual data labeling and classification. Therefore, personal subjective factors will be incorporated into the classification. With the increasing number of enterprises, there will be problems such as high labeling cost and incomplete coverage. . At this stage, some traditional machine learning methods are used, but the traditional classification methods are not satisfactory, and the generalization ability is not strong. There are also a small number of people who use deep learning classification methods, but only ...

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): G06F16/35G06F40/211G06F40/30G06N3/04G06N3/08
CPCG06F16/35G06F40/211G06F40/30G06N3/08G06F2216/03G06N3/048G06N3/044G06N3/045
Inventor 刘祥志于洋吴晓明石金泽薛许强张鹏汪付强张建强郝秋赟马晓凤满佳政孙丰收乔友为
Owner SHANDONG COMP SCI CENTNAT SUPERCOMP CENT IN JINAN
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