Enterprise industry secondary industry multi-label classifier based on deep learning algorithm

A deep learning, multi-label technology, applied in special data processing applications, biological neural network models, structured data retrieval, etc., can solve the problems of low accuracy, difficult maintenance, troublesome and other problems, and improve the training time is too long. , the effect of improving the accuracy

Pending Publication Date: 2021-09-17
国科元科技(北京)有限公司
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide an enterprise industry secondary industry multi-label classifier based on a deep learning algorithm to solve the problem that the existing technology proposed in the above background technology does not support the enterprise secondary industry classification and multi-label classification. The accuracy rate of the method used in the existing technology is not high, it is difficult to complete the multi-label classification task in the existing technology, and the existing technology is cumbersome, complicated, troublesome and difficult to maintain

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 industry secondary industry multi-label classifier based on deep learning algorithm
  • Enterprise industry secondary industry multi-label classifier based on deep learning algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] see figure 1 and figure 2 , the present invention provides a technical solution: a multi-label classifier for enterprise industry secondary industry based on deep learning algorithm, which consists of a collection module, a preprocessing module, a management module, a model building module, a training verification module, an input module, and a display module. consists of:

[0040] The collection module is used to collect the business scope information of the enterprise;

[0041] The preprocessing module is used to preprocess the business scope information of the enterprise;

[0042] The management module is used to manually index the business scope information of the enterprise, and to produce training sets, validation sets and test sets for multi-label classification training;

[0043] The model building module is used to build the Albert+TextCNN model using the training set;

[0044] The training and verification module is used to train the established Albert+Te...

Embodiment 2

[0070] see figure 1 and figure 2 , the present invention provides a technical solution: a multi-label classifier for enterprise industry secondary industry based on deep learning algorithm, which consists of a collection module, a preprocessing module, a management module, a model building module, a training verification module, an input module, and a display module. consists of:

[0071] The collection module is used to collect the business scope information of the enterprise;

[0072] The preprocessing module is used to preprocess the business scope information of the enterprise;

[0073] The management module is used to manually index the business scope information of the enterprise, and to produce training sets, validation sets and test sets for multi-label classification training;

[0074] The model building module is used to build the Albert+TextCNN model using the training set;

[0075] The training and verification module is used to train the established Albert+Te...

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 enterprise industry secondary industry multi-label classifier based on a deep learning algorithm, which is composed of an acquisition module, a preprocessing module, a management module, a model establishment module, a training verification module, an input module and a display module. The acquisition module is used for acquiring enterprise operation scope information. The preprocessing module is used for preprocessing the enterprise operation scope information. The management module is used for manually indexing the enterprise operation scope information and making a training set, a verification set and a test set for multi-label classification training. The model establishment module is used for establishing an Albert + TextCNN model by using the training set. The training verification module is used for training the established Albert + TextCNN model and verifying an accuracy rate. The method has the beneficial effects that the multi-label classification of the secondary industry of the enterprise is realized, and the problem that the training time required by an existing method is too long is improved; and through modular design, single-label classification can be carried out, and the accuracy is greatly improved compared with that of the existing method.

Description

technical field [0001] The invention belongs to the technical field of NLP natural language processing, and in particular relates to a multi-label classifier for enterprise industry secondary industries based on a deep learning algorithm. Background technique [0002] NLP Natural Language Processing refers to developing applications or services that can understand human language. [0003] Existing technologies are mainly divided into three categories, namely unsupervised and semi-supervised learning methods, traditional machine learning methods and deep learning methods; among them: unsupervised learning methods and semi-supervised learning methods are usually artificially formulated standards and Extract features, such as using term frequency (TF), inverse document frequency (IDF), logistic regression (Logistic Regression), decision tree (Decision Tree), mutual information (Mutual Information), k-adjacent value (K-adjacent value), adaptive Enhancement (AdaBoost) and multi-...

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): G06F16/28G06N3/04
CPCG06F16/285G06N3/045
Inventor 陈鹏王树志梁正尧马金河
Owner 国科元科技(北京)有限公司
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